fix: use -p never shorthand for electron-builder

The --publish never format wasn't being parsed correctly. Using -p never shorthand instead.

.design-system/src/lib/icons.ts

@@ -0,0 +1,39 @@
+/**
+ * Centralized Icon Exports for Design System
+ *
+ * This file serves as the single source of truth for all lucide-react icons used
+ * throughout the design system demo app. By consolidating imports here, we enable:
+ *
+ * 1. Better tracking of which icons are actually used
+ * 2. Potential code-splitting opportunities
+ * 3. Easier future migration to alternative icon solutions
+ * 4. Reduced bundle size through optimized tree-shaking
+ *
+ * Usage:
+ *   import { Check, ChevronLeft, X } from '../lib/icons';
+ *
+ * When adding new icons:
+ *   1. Import the icon from 'lucide-react'
+ *   2. Add it to the export statement in alphabetical order
+ */
+
+export {
+  Check,
+  ChevronLeft,
+  ChevronRight,
+  Github,
+  Heart,
+  MessageSquare,
+  Minus,
+  Moon,
+  MoreVertical,
+  Plus,
+  RotateCcw,
+  Slack,
+  Sparkles,
+  Star,
+  Sun,
+  Video,
+  X,
+  Zap,
+} from 'lucide-react';

.github/release-drafter.yml

@@ -0,0 +1,35 @@
+name-template: 'v$RESOLVED_VERSION'
+tag-template: 'v$RESOLVED_VERSION'
+
+categories:
+  - title: '## New Features'
+    labels:
+      - 'feature'
+      - 'enhancement'
+  - title: '## Bug Fixes'
+    labels:
+      - 'bug'
+      - 'fix'
+  - title: '## Improvements'
+    labels:
+      - 'improvement'
+      - 'refactor'
+  - title: '## Documentation'
+    labels:
+      - 'documentation'
+  - title: '## Other Changes'
+    labels:
+      - '*'
+
+change-template: '* $TITLE (#$NUMBER) @$AUTHOR'
+
+sort-by: merged_at
+sort-direction: ascending
+
+template: |
+  $CHANGES
+
+  **Full Changelog**: https://github.com/$OWNER/$REPOSITORY/compare/$PREVIOUS_TAG...$RESOLVED_VERSION
+
+  ## Contributors
+  $CONTRIBUTORS

.github/workflows/ci.yml

@@ -57,7 +57,7 @@ jobs:
         env:
           CODECOV_TOKEN: ${{ secrets.CODECOV_TOKEN }}
 
-  # Frontend tests
+  # Frontend lint, typecheck, test, and build
   test-frontend:
     runs-on: ubuntu-latest
     steps:
@@ -72,12 +72,34 @@ jobs:
       - name: Setup pnpm
         uses: pnpm/action-setup@v4
         with:
-          version: 9
+          version: 10
+
+      - name: Get pnpm store directory
+        id: pnpm-cache
+        run: echo "dir=$(pnpm store path)" >> "$GITHUB_OUTPUT"
+
+      - uses: actions/cache@v4
+        with:
+          path: ${{ steps.pnpm-cache.outputs.dir }}
+          key: ${{ runner.os }}-pnpm-${{ hashFiles('**/pnpm-lock.yaml') }}
+          restore-keys: ${{ runner.os }}-pnpm-
 
       - name: Install dependencies
         working-directory: auto-claude-ui
         run: pnpm install --frozen-lockfile --ignore-scripts
 
+      - name: Lint
+        working-directory: auto-claude-ui
+        run: pnpm run lint
+
+      - name: Type check
+        working-directory: auto-claude-ui
+        run: pnpm run typecheck
+
       - name: Run tests
         working-directory: auto-claude-ui
-        run: pnpm test
+        run: pnpm run test
+
+      - name: Build
+        working-directory: auto-claude-ui
+        run: pnpm run build

.github/workflows/discord-release.yml

@@ -22,4 +22,3 @@ jobs:
           footer_timestamp: true
           reduce_headings: true
           remove_github_reference_links: true
-

.github/workflows/release.yml

@@ -0,0 +1,453 @@
+name: Release
+
+on:
+  push:
+    tags:
+      - 'v*'
+  workflow_dispatch:
+    inputs:
+      dry_run:
+        description: 'Test build without creating release'
+        required: false
+        default: true
+        type: boolean
+
+jobs:
+  # Intel build on Intel runner for native compilation
+  # Note: macos-15-intel is the last Intel runner, supported until Fall 2027
+  build-macos-intel:
+    runs-on: macos-15-intel
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version: '20'
+
+      - name: Setup pnpm
+        uses: pnpm/action-setup@v4
+        with:
+          version: 10
+
+      - name: Get pnpm store directory
+        id: pnpm-cache
+        run: echo "dir=$(pnpm store path)" >> $GITHUB_OUTPUT
+
+      - uses: actions/cache@v4
+        with:
+          path: ${{ steps.pnpm-cache.outputs.dir }}
+          key: ${{ runner.os }}-x64-pnpm-${{ hashFiles('**/pnpm-lock.yaml') }}
+          restore-keys: ${{ runner.os }}-x64-pnpm-
+
+      - name: Install dependencies
+        run: cd auto-claude-ui && pnpm install --frozen-lockfile
+
+      - name: Build application
+        run: cd auto-claude-ui && pnpm run build
+
+      - name: Package macOS (Intel)
+        run: cd auto-claude-ui && pnpm run package:mac -- --arch=x64 -p never
+        env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          CSC_LINK: ${{ secrets.MAC_CERTIFICATE }}
+          CSC_KEY_PASSWORD: ${{ secrets.MAC_CERTIFICATE_PASSWORD }}
+
+      - name: Notarize macOS Intel app
+        env:
+          APPLE_ID: ${{ secrets.APPLE_ID }}
+          APPLE_APP_SPECIFIC_PASSWORD: ${{ secrets.APPLE_APP_SPECIFIC_PASSWORD }}
+          APPLE_TEAM_ID: ${{ secrets.APPLE_TEAM_ID }}
+        run: |
+          if [ -z "$APPLE_ID" ]; then
+            echo "Skipping notarization: APPLE_ID not configured"
+            exit 0
+          fi
+          cd auto-claude-ui
+          for dmg in dist/*.dmg; do
+            echo "Notarizing $dmg..."
+            xcrun notarytool submit "$dmg" \
+              --apple-id "$APPLE_ID" \
+              --password "$APPLE_APP_SPECIFIC_PASSWORD" \
+              --team-id "$APPLE_TEAM_ID" \
+              --wait
+            xcrun stapler staple "$dmg"
+            echo "Successfully notarized and stapled $dmg"
+          done
+
+      - name: Upload artifacts
+        uses: actions/upload-artifact@v4
+        with:
+          name: macos-intel-builds
+          path: |
+            auto-claude-ui/dist/*.dmg
+            auto-claude-ui/dist/*.zip
+
+  # Apple Silicon build on ARM64 runner for native compilation
+  build-macos-arm64:
+    runs-on: macos-15
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version: '20'
+
+      - name: Setup pnpm
+        uses: pnpm/action-setup@v4
+        with:
+          version: 10
+
+      - name: Get pnpm store directory
+        id: pnpm-cache
+        run: echo "dir=$(pnpm store path)" >> $GITHUB_OUTPUT
+
+      - uses: actions/cache@v4
+        with:
+          path: ${{ steps.pnpm-cache.outputs.dir }}
+          key: ${{ runner.os }}-arm64-pnpm-${{ hashFiles('**/pnpm-lock.yaml') }}
+          restore-keys: ${{ runner.os }}-arm64-pnpm-
+
+      - name: Install dependencies
+        run: cd auto-claude-ui && pnpm install --frozen-lockfile
+
+      - name: Build application
+        run: cd auto-claude-ui && pnpm run build
+
+      - name: Package macOS (Apple Silicon)
+        run: cd auto-claude-ui && pnpm run package:mac -- --arch=arm64 -p never
+        env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          CSC_LINK: ${{ secrets.MAC_CERTIFICATE }}
+          CSC_KEY_PASSWORD: ${{ secrets.MAC_CERTIFICATE_PASSWORD }}
+
+      - name: Notarize macOS ARM64 app
+        env:
+          APPLE_ID: ${{ secrets.APPLE_ID }}
+          APPLE_APP_SPECIFIC_PASSWORD: ${{ secrets.APPLE_APP_SPECIFIC_PASSWORD }}
+          APPLE_TEAM_ID: ${{ secrets.APPLE_TEAM_ID }}
+        run: |
+          if [ -z "$APPLE_ID" ]; then
+            echo "Skipping notarization: APPLE_ID not configured"
+            exit 0
+          fi
+          cd auto-claude-ui
+          for dmg in dist/*.dmg; do
+            echo "Notarizing $dmg..."
+            xcrun notarytool submit "$dmg" \
+              --apple-id "$APPLE_ID" \
+              --password "$APPLE_APP_SPECIFIC_PASSWORD" \
+              --team-id "$APPLE_TEAM_ID" \
+              --wait
+            xcrun stapler staple "$dmg"
+            echo "Successfully notarized and stapled $dmg"
+          done
+
+      - name: Upload artifacts
+        uses: actions/upload-artifact@v4
+        with:
+          name: macos-arm64-builds
+          path: |
+            auto-claude-ui/dist/*.dmg
+            auto-claude-ui/dist/*.zip
+
+  build-windows:
+    runs-on: windows-latest
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version: '20'
+
+      - name: Setup pnpm
+        uses: pnpm/action-setup@v4
+        with:
+          version: 10
+
+      - name: Get pnpm store directory
+        id: pnpm-cache
+        shell: bash
+        run: echo "dir=$(pnpm store path)" >> $GITHUB_OUTPUT
+
+      - uses: actions/cache@v4
+        with:
+          path: ${{ steps.pnpm-cache.outputs.dir }}
+          key: ${{ runner.os }}-pnpm-${{ hashFiles('**/pnpm-lock.yaml') }}
+          restore-keys: ${{ runner.os }}-pnpm-
+
+      - name: Install dependencies
+        run: cd auto-claude-ui && pnpm install --frozen-lockfile
+
+      - name: Build application
+        run: cd auto-claude-ui && pnpm run build
+
+      - name: Package Windows
+        run: cd auto-claude-ui && pnpm run package:win -- -p never
+        env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          CSC_LINK: ${{ secrets.WIN_CERTIFICATE }}
+          CSC_KEY_PASSWORD: ${{ secrets.WIN_CERTIFICATE_PASSWORD }}
+
+      - name: Upload artifacts
+        uses: actions/upload-artifact@v4
+        with:
+          name: windows-builds
+          path: |
+            auto-claude-ui/dist/*.exe
+
+  build-linux:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version: '20'
+
+      - name: Setup pnpm
+        uses: pnpm/action-setup@v4
+        with:
+          version: 10
+
+      - name: Get pnpm store directory
+        id: pnpm-cache
+        run: echo "dir=$(pnpm store path)" >> $GITHUB_OUTPUT
+
+      - uses: actions/cache@v4
+        with:
+          path: ${{ steps.pnpm-cache.outputs.dir }}
+          key: ${{ runner.os }}-pnpm-${{ hashFiles('**/pnpm-lock.yaml') }}
+          restore-keys: ${{ runner.os }}-pnpm-
+
+      - name: Install dependencies
+        run: cd auto-claude-ui && pnpm install --frozen-lockfile
+
+      - name: Build application
+        run: cd auto-claude-ui && pnpm run build
+
+      - name: Package Linux
+        run: cd auto-claude-ui && pnpm run package:linux -- -p never
+        env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+
+      - name: Upload artifacts
+        uses: actions/upload-artifact@v4
+        with:
+          name: linux-builds
+          path: |
+            auto-claude-ui/dist/*.AppImage
+            auto-claude-ui/dist/*.deb
+
+  create-release:
+    needs: [build-macos-intel, build-macos-arm64, build-windows, build-linux]
+    runs-on: ubuntu-latest
+    permissions:
+      contents: write
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - name: Download all artifacts
+        uses: actions/download-artifact@v4
+        with:
+          path: dist
+
+      - name: Flatten and validate artifacts
+        run: |
+          mkdir -p release-assets
+          find dist -type f \( -name "*.dmg" -o -name "*.zip" -o -name "*.exe" -o -name "*.AppImage" -o -name "*.deb" \) -exec cp {} release-assets/ \;
+
+          # Validate that at least one artifact was copied
+          artifact_count=$(find release-assets -type f \( -name "*.dmg" -o -name "*.zip" -o -name "*.exe" -o -name "*.AppImage" -o -name "*.deb" \) | wc -l)
+          if [ "$artifact_count" -eq 0 ]; then
+            echo "::error::No build artifacts found! Expected .dmg, .zip, .exe, .AppImage, or .deb files."
+            exit 1
+          fi
+
+          echo "Found $artifact_count artifact(s):"
+          ls -la release-assets/
+
+      - name: Generate checksums
+        run: |
+          cd release-assets
+          sha256sum ./* > checksums.sha256
+          cat checksums.sha256
+
+      - name: Scan with VirusTotal
+        id: virustotal
+        continue-on-error: true
+        if: ${{ github.event_name == 'push' || (github.event_name == 'workflow_dispatch' && inputs.dry_run != true) }}
+        env:
+          VT_API_KEY: ${{ secrets.VIRUSTOTAL_API_KEY }}
+        run: |
+          if [ -z "$VT_API_KEY" ]; then
+            echo "::warning::VIRUSTOTAL_API_KEY not configured, skipping scan"
+            echo "vt_results=" >> $GITHUB_OUTPUT
+            exit 0
+          fi
+
+          echo "## VirusTotal Scan Results" > vt_results.md
+          echo "" >> vt_results.md
+
+          for file in release-assets/*.{exe,dmg,AppImage,deb}; do
+            [ -f "$file" ] || continue
+            filename=$(basename "$file")
+            filesize=$(stat -c%s "$file" 2>/dev/null || stat -f%z "$file")
+            echo "Scanning $filename (${filesize} bytes)..."
+
+            # For files > 32MB, get a special upload URL first
+            if [ "$filesize" -gt 33554432 ]; then
+              echo "  Large file detected, requesting upload URL..."
+              upload_url_response=$(curl -s --request GET \
+                --url "https://www.virustotal.com/api/v3/files/upload_url" \
+                --header "x-apikey: $VT_API_KEY")
+
+              upload_url=$(echo "$upload_url_response" | jq -r '.data // empty')
+              if [ -z "$upload_url" ]; then
+                echo "::warning::Failed to get upload URL for large file $filename"
+                echo "Response: $upload_url_response"
+                echo "- $filename - ⚠️ Upload failed (large file)" >> vt_results.md
+                continue
+              fi
+              api_url="$upload_url"
+            else
+              api_url="https://www.virustotal.com/api/v3/files"
+            fi
+
+            # Upload file to VirusTotal
+            response=$(curl -s --request POST \
+              --url "$api_url" \
+              --header "x-apikey: $VT_API_KEY" \
+              --form "file=@$file")
+
+            # Check if response is valid JSON before parsing
+            if ! echo "$response" | jq -e . >/dev/null 2>&1; then
+              echo "::warning::VirusTotal returned invalid JSON for $filename"
+              echo "Response (first 500 chars): ${response:0:500}"
+              echo "- $filename - ⚠️ Scan failed (invalid response)" >> vt_results.md
+              continue
+            fi
+
+            # Check for API error response
+            error_code=$(echo "$response" | jq -r '.error.code // empty')
+            if [ -n "$error_code" ]; then
+              error_msg=$(echo "$response" | jq -r '.error.message // "Unknown error"')
+              echo "::warning::VirusTotal API error for $filename: $error_code - $error_msg"
+              echo "- $filename - ⚠️ Scan failed ($error_code)" >> vt_results.md
+              continue
+            fi
+
+            # Extract analysis ID
+            analysis_id=$(echo "$response" | jq -r '.data.id // empty')
+
+            if [ -z "$analysis_id" ]; then
+              echo "::warning::Failed to upload $filename to VirusTotal"
+              echo "Response: $response"
+              echo "- $filename - ⚠️ Upload failed" >> vt_results.md
+              continue
+            fi
+
+            echo "Uploaded $filename, analysis ID: $analysis_id"
+
+            # Wait for analysis to complete (max 5 minutes per file)
+            analysis=""
+            for i in {1..30}; do
+              sleep 10
+              analysis=$(curl -s --request GET \
+                --url "https://www.virustotal.com/api/v3/analyses/$analysis_id" \
+                --header "x-apikey: $VT_API_KEY")
+
+              # Validate JSON response
+              if ! echo "$analysis" | jq -e . >/dev/null 2>&1; then
+                echo "  Warning: Invalid JSON response on attempt $i, retrying..."
+                continue
+              fi
+
+              status=$(echo "$analysis" | jq -r '.data.attributes.status // "unknown"')
+              echo "  Status: $status (attempt $i/30)"
+
+              if [ "$status" = "completed" ]; then
+                break
+              fi
+            done
+
+            # Final validation that we have valid analysis data
+            if ! echo "$analysis" | jq -e '.data.attributes.stats' >/dev/null 2>&1; then
+              echo "::warning::Could not get complete analysis for $filename, using local hash"
+              file_hash=$(sha256sum "$file" | cut -d' ' -f1)
+              echo "- [$filename](https://www.virustotal.com/gui/file/$file_hash) - ⚠️ Analysis incomplete" >> vt_results.md
+              continue
+            fi
+
+            # Get file hash for permanent URL
+            file_hash=$(echo "$analysis" | jq -r '.meta.file_info.sha256 // empty')
+
+            if [ -z "$file_hash" ]; then
+              # Fallback: calculate hash locally
+              file_hash=$(sha256sum "$file" | cut -d' ' -f1)
+            fi
+
+            # Get detection stats
+            malicious=$(echo "$analysis" | jq -r '.data.attributes.stats.malicious // 0')
+            suspicious=$(echo "$analysis" | jq -r '.data.attributes.stats.suspicious // 0')
+            undetected=$(echo "$analysis" | jq -r '.data.attributes.stats.undetected // 0')
+
+            vt_url="https://www.virustotal.com/gui/file/$file_hash"
+
+            if [ "$malicious" -gt 0 ] || [ "$suspicious" -gt 0 ]; then
+              echo "::warning::$filename has $malicious malicious and $suspicious suspicious detections (likely false positives)"
+              echo "- [$filename]($vt_url) - ⚠️ **$malicious malicious, $suspicious suspicious** detections (review recommended)" >> vt_results.md
+            else
+              echo "$filename is clean ($undetected engines, 0 detections)"
+              echo "- [$filename]($vt_url) - ✅ Clean ($undetected engines, 0 detections)" >> vt_results.md
+            fi
+          done
+
+          echo "" >> vt_results.md
+
+          # Save results for release notes
+          cat vt_results.md
+          echo "vt_results<<EOF" >> $GITHUB_OUTPUT
+          cat vt_results.md >> $GITHUB_OUTPUT
+          echo "EOF" >> $GITHUB_OUTPUT
+
+      - name: Dry run summary
+        if: ${{ github.event_name == 'workflow_dispatch' && inputs.dry_run == true }}
+        run: |
+          echo "## Dry Run Complete" >> $GITHUB_STEP_SUMMARY
+          echo "Build artifacts created successfully:" >> $GITHUB_STEP_SUMMARY
+          echo "\`\`\`" >> $GITHUB_STEP_SUMMARY
+          ls -la release-assets/ >> $GITHUB_STEP_SUMMARY
+          echo "\`\`\`" >> $GITHUB_STEP_SUMMARY
+          echo "### Checksums" >> $GITHUB_STEP_SUMMARY
+          echo "\`\`\`" >> $GITHUB_STEP_SUMMARY
+          cat release-assets/checksums.sha256 >> $GITHUB_STEP_SUMMARY
+          echo "\`\`\`" >> $GITHUB_STEP_SUMMARY
+
+      - name: Generate changelog
+        if: ${{ github.event_name == 'push' || (github.event_name == 'workflow_dispatch' && inputs.dry_run != true) }}
+        id: changelog
+        uses: release-drafter/release-drafter@v6
+        with:
+          config-name: release-drafter.yml
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+
+      - name: Create Release
+        if: ${{ github.event_name == 'push' || (github.event_name == 'workflow_dispatch' && inputs.dry_run != true) }}
+        uses: softprops/action-gh-release@v2
+        with:
+          body: |
+            ${{ steps.changelog.outputs.body }}
+
+            ${{ steps.virustotal.outputs.vt_results }}
+          files: release-assets/*
+          draft: false
+          prerelease: ${{ contains(github.ref, 'beta') || contains(github.ref, 'alpha') }}
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

.github/workflows/validate-version.yml

@@ -0,0 +1,71 @@
+name: Validate Version
+
+on:
+  push:
+    tags:
+      - 'v*'
+
+jobs:
+  validate-version:
+    name: Validate package.json version matches tag
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Extract version from tag
+        id: tag_version
+        run: |
+          # Extract version from tag (e.g., v2.5.5 -> 2.5.5)
+          TAG_VERSION=${GITHUB_REF#refs/tags/v}
+          echo "version=$TAG_VERSION" >> $GITHUB_OUTPUT
+          echo "Tag version: $TAG_VERSION"
+
+      - name: Extract version from package.json
+        id: package_version
+        run: |
+          # Read version from package.json
+          PACKAGE_VERSION=$(node -p "require('./auto-claude-ui/package.json').version")
+          echo "version=$PACKAGE_VERSION" >> $GITHUB_OUTPUT
+          echo "Package.json version: $PACKAGE_VERSION"
+
+      - name: Compare versions
+        run: |
+          TAG_VERSION="${{ steps.tag_version.outputs.version }}"
+          PACKAGE_VERSION="${{ steps.package_version.outputs.version }}"
+
+          echo "=========================================="
+          echo "Version Validation"
+          echo "=========================================="
+          echo "Git tag version:      v$TAG_VERSION"
+          echo "package.json version: $PACKAGE_VERSION"
+          echo "=========================================="
+
+          if [ "$TAG_VERSION" != "$PACKAGE_VERSION" ]; then
+            echo ""
+            echo "❌ ERROR: Version mismatch detected!"
+            echo ""
+            echo "The version in package.json ($PACKAGE_VERSION) does not match"
+            echo "the git tag version ($TAG_VERSION)."
+            echo ""
+            echo "To fix this:"
+            echo "  1. Delete this tag: git tag -d v$TAG_VERSION"
+            echo "  2. Update package.json version to $TAG_VERSION"
+            echo "  3. Commit the change"
+            echo "  4. Recreate the tag: git tag -a v$TAG_VERSION -m 'Release v$TAG_VERSION'"
+            echo ""
+            echo "Or use the automated script:"
+            echo "  node scripts/bump-version.js $TAG_VERSION"
+            echo ""
+            exit 1
+          fi
+
+          echo ""
+          echo "✅ SUCCESS: Versions match!"
+          echo ""
+
+      - name: Version validation result
+        if: success()
+        run: |
+          echo "::notice::Version validation passed - package.json version matches tag v${{ steps.tag_version.outputs.version }}"

.gitignore

@@ -86,4 +86,4 @@ dev/
 _bmad
 _bmad-output
 
-.claude
+.claude

.pre-commit-config.yaml

@@ -33,4 +33,5 @@ repos:
       - id: trailing-whitespace
       - id: end-of-file-fixer
       - id: check-yaml
+        exclude: pnpm-lock\.yaml$
       - id: check-added-large-files

CHANGELOG.md

@@ -1,3 +1,446 @@
+## 2.7.1 - Build Pipeline Enhancements
+
+### 🛠️ Improvements
+
+- Enhanced VirusTotal scan error handling in release workflow with graceful failure recovery and improved reporting visibility
+
+- Refactored macOS build workflow to support both Intel and ARM64 architectures with notarization for Intel builds and improved artifact handling
+
+- Streamlined CI/CD processes with updated caching strategies and enhanced error handling for external API interactions
+
+### 📚 Documentation
+
+- Clarified README documentation
+
+---
+
+## What's Changed
+
+- chore: Enhance VirusTotal scan error handling in release workflow by @AndyMik90 in d23fcd8
+
+- chore: Refactor macOS build workflow to support Intel and ARM64 architectures by @AndyMik90 in 326118b
+
+- docs: readme clarification by @AndyMik90 in 6afcc92
+
+- fix: version by @AndyMik90 in 2c93890
+
+## Thanks to all contributors
+
+@AndyMik90
+
+## 2.7.0 - Tab Persistence & Memory System Modernization
+
+### ✨ New Features
+
+- Project tab bar with persistent tab management and GitHub organization initialization on project creation
+
+- Task creation enhanced with @ autocomplete for agent profiles and improved drag-and-drop support
+
+- Keyboard shortcuts and tooltips added to project tabs for better navigation
+
+- Agent task restart functionality with new profile support for flexible task recovery
+
+- Ollama embedding model support with automatic dimension detection for self-hosted deployments
+
+### 🛠️ Improvements
+
+- Memory system completely redesigned with embedded LadybugDB, eliminating Docker/FalkorDB dependency and improving performance
+
+- Tab persistence implemented via IPC-based mechanism for reliable session state management
+
+- Terminal environment improved by using virtual environment Python for proper terminal name generation
+
+- AI merge operations timeout increased from 2 to 10 minutes for reliability with larger changes
+
+- Merge operations now use stored baseBranch metadata for consistent branch targeting
+
+- Memory configuration UI simplified and rebranded with improved Ollama integration and detection
+
+- CI/CD workflows enhanced with code signing support and automated release process
+
+- Cross-platform compatibility improved by replacing Unix shell syntax with portable git commands
+
+- Python venv created in userData for packaged applications to ensure proper environment isolation
+
+### 🐛 Bug Fixes
+
+- Task title no longer blocks edit/close buttons in UI
+
+- Tab persistence and terminal shortcuts properly scoped to prevent conflicts
+
+- Agent profile fallback corrected from 'Balanced' to 'Auto (Optimized)'
+
+- macOS notarization made optional and improved with private artifact storage
+
+- Embedding provider changes now properly detected during migration
+
+- Memory query CLI respects user's memory enabled flag
+
+- CodeRabbit review issues and linting errors resolved across codebase
+
+- F-string prefixes removed from strings without placeholders
+
+- Import ordering fixed for ruff compliance
+
+- Preview panel now receives projectPath prop correctly for image component functionality
+
+- Default database path unified to ~/.auto-claude/memories for consistency
+
+- @lydell/node-pty build scripts compatibility improved for pnpm v10
+
+---
+
+## What's Changed
+
+- feat(ui): add project tab bar from PR #101 by @AndyMik90 in c400fe9
+
+- feat: improve task creation UX with @ autocomplete and better drag-drop by @AndyMik90 in 20d1487
+
+- feat(ui): add keyboard shortcuts and tooltips for project tabs by @AndyMik90 in ed73265
+
+- feat(agent): enhance task restart functionality with new profile support by @AndyMik90 in c8452a5
+
+- feat: add Ollama embedding model support with auto-detected dimensions by @AndyMik90 in 45901f3
+
+- feat(memory): replace FalkorDB with LadybugDB embedded database by @AndyMik90 in 87d0b52
+
+- feat: add automated release workflow with code signing by @AndyMik90 in 6819b00
+
+- feat: add embedding provider change detection and fix import ordering by @AndyMik90 in 36f8006
+
+- fix(tests): update tab management tests for IPC-based persistence by @AndyMik90 in ea25d6e
+
+- fix(ui): address CodeRabbit PR review issues by @AndyMik90 in 39ce754
+
+- fix: address CodeRabbit review issues by @AndyMik90 in 95ae0b0
+
+- fix: prevent task title from blocking edit/close buttons by @AndyMik90 in 8a0fb26
+
+- fix: use venv Python for terminal name generation by @AndyMik90 in 325cb54
+
+- fix(merge): increase AI merge timeout from 2 to 10 minutes by @AndyMik90 in 4477538
+
+- fix(merge): use stored baseBranch from task metadata for merge operations by @AndyMik90 in 8d56474
+
+- fix: unify default database path to ~/.auto-claude/memories by @AndyMik90 in 684e3f9
+
+- fix(ui): fix tab persistence and scope terminal shortcuts by @AndyMik90 in 2d1168b
+
+- fix: create Python venv in userData for packaged apps by @AndyMik90 in b83377c
+
+- fix(ui): change agent profile fallback from 'Balanced' to 'Auto (Optimized)' by @AndyMik90 in 385dcc1
+
+- fix: check APPLE_ID in shell instead of workflow if condition by @AndyMik90 in 9eece01
+
+- fix: allow @lydell/node-pty build scripts in pnpm v10 by @AndyMik90 in 1f6963f
+
+- fix: use shell guard for notarization credentials check by @AndyMik90 in 4cbddd3
+
+- fix: improve migrate_embeddings robustness and correctness by @AndyMik90 in 61f0238
+
+- fix: respect user's memory enabled flag in query_memory CLI by @AndyMik90 in 45b2c83
+
+- fix: save notarization logs to private artifact instead of public logs by @AndyMik90 in a82525d
+
+- fix: make macOS notarization optional by @AndyMik90 in f2b7b56
+
+- fix: add author email for Linux builds by @AndyMik90 in 5f66127
+
+- fix: add GH_TOKEN and homepage for release workflow by @AndyMik90 in 568ea18
+
+- fix(ci): quote GITHUB_OUTPUT for shell safety by @AndyMik90 in 1e891e1
+
+- fix: address CodeRabbit review feedback by @AndyMik90 in 8e4b1da
+
+- fix: update test and apply ruff formatting by @AndyMik90 in a087ba3
+
+- fix: address additional CodeRabbit review comments by @AndyMik90 in 461fad6
+
+- fix: sort imports in memory.py for ruff I001 by @AndyMik90 in b3c257d
+
+- fix: address CodeRabbit review comments from PR #100 by @AndyMik90 in 1ed237a
+
+- fix: remove f-string prefixes from strings without placeholders by @AndyMik90 in bcd453a
+
+- fix: resolve remaining CI failures by @AndyMik90 in cfbccda
+
+- fix: resolve all CI failures in PR #100 by @AndyMik90 in c493d6c
+
+- fix(cli): update graphiti status display for LadybugDB by @AndyMik90 in 049c60c
+
+- fix(ui): replace Unix shell syntax with cross-platform git commands by @AndyMik90 in 83aa3f0
+
+- fix: correct model name and release workflow conditionals by @AndyMik90 in de41dfc
+
+- style: fix ruff linting errors in graphiti queries by @AndyMik90 in 127559f
+
+- style: apply ruff formatting to 4 files by @AndyMik90 in 9d5d075
+
+- refactor: update memory test suite for LadybugDB by @AndyMik90 in f0b5efc
+
+- refactor(ui): simplify reference files and images handling in task modal by @AndyMik90 in 1975e4d
+
+- refactor: rebrand memory system UI and simplify configuration by @AndyMik90 in 2b3cd49
+
+- refactor: replace Docker/FalkorDB with embedded LadybugDB for memory system by @AndyMik90 in 325458d
+
+- docs: add CodeRabbit review response tracking by @AndyMik90 in 3452548
+
+- chore: use GitHub noreply email for author field by @AndyMik90 in 18f2045
+
+- chore: simplify notarization step after successful setup by @AndyMik90 in e4fe7cd
+
+- chore: update CI and release workflows, remove changelog config by @AndyMik90 in 6f891b7
+
+- chore: remove docker-compose.yml (FalkorDB no longer used) by @AndyMik90 in 68f3f06
+
+- fix: Replace space with hyphen in productName to fix PTY daemon spawn (#65) by @Craig Van in 8f1f7a7
+
+- fix: update npm scripts to use hyphenated product name by @AndyMik90 in 89978ed
+
+- fix(ui): improve Ollama UX in memory settings by @AndyMik90 in dea1711
+
+- auto-claude: subtask-1-1 - Add projectPath prop to PreviewPanel and implement custom img component by @AndyMik90 in e6529e0
+
+- Project tab persistence and github org init on project creation by @AndyMik90 in ae1dac9
+
+- Readme for installors by @AndyMik90 in 1855d7d
+
+---
+
+## Thanks to all contributors
+
+@AndyMik90, @Craig Van
+
+## 2.6.0 - Improved User Experience and Agent Configuration
+
+### ✨ New Features
+
+- Add customizable phase configuration in app settings, allowing users to tailor the AI build pipeline to their workflow
+
+- Implement parallel AI merge functionality for faster integration of completed builds
+
+- Add Google AI as LLM and embedding provider for Graphiti memory system
+
+- Implement device code authentication flow with timeout handling, browser launch fallback, and comprehensive testing
+
+### 🛠️ Improvements
+
+- Move Agent Profiles from dashboard to Settings for better organization and discoverability
+
+- Default agent profile to 'Auto (Optimized)' for streamlined out-of-the-box experience
+
+- Enhance WorkspaceStatus component UI with improved visual design
+
+- Refactor task management from sidebar to modal interface for cleaner navigation
+
+- Add comprehensive theme system with multiple color schemes (Forest, Neo, Retro, Dusk, Ocean, Lime) and light/dark mode support
+
+- Extract human-readable feature titles from spec.md for better task identification
+
+- Improve task description display for specs with compact markdown formatting
+
+### 🐛 Bug Fixes
+
+- Fix asyncio coroutine creation in worker threads to properly support async operations
+
+- Improve UX for phase configuration in task creation workflow
+
+- Address CodeRabbit PR #69 feedback and additional review comments
+
+- Fix auto-close behavior for task modal when marking tasks as done
+
+- Resolve Python lint errors and import sorting issues (ruff I001 compliance)
+
+- Ensure planner agent properly writes implementation_plan.json
+
+- Add platform detection for terminal profile commands on Windows
+
+- Set default selected agent profile to 'auto' across all users
+
+- Fix display of correct merge target branch in worktree UI
+
+- Add validation for invalid colorTheme fallback to prevent UI errors
+
+- Remove outdated Sun/Moon toggle button from sidebar
+
+---
+
+## What's Changed
+
+- feat: add customizable phase configuration in app settings by @AndyMik90 in aee0ba4
+
+- feat: implement parallel AI merge functionality by @AndyMik90 in 458d4bb
+
+- feat(graphiti): add Google AI as LLM and embedding provider by @adryserage in fe69106
+
+- fix: create coroutine inside worker thread for asyncio.run by @AndyMik90 in f89e4e6
+
+- fix: improve UX for phase configuration in task creation by @AndyMik90 in b9797cb
+
+- fix: address CodeRabbit PR #69 feedback by @AndyMik90 in cc38a06
+
+- fix: sort imports in workspace.py to pass ruff I001 check by @AndyMik90 in 9981ee4
+
+- fix(ui): auto-close task modal when marking task as done by @AndyMik90 in 297d380
+
+- fix: resolve Python lint errors in workspace.py by @AndyMik90 in 0506256
+
+- refactor: move Agent Profiles from dashboard to Settings by @AndyMik90 in 1094990
+
+- fix(planning): ensure planner agent writes implementation_plan.json by @AndyMik90 in 9ab5a4f
+
+- fix(windows): add platform detection for terminal profile commands by @AndyMik90 in f0a6a0a
+
+- fix: default agent profile to 'Auto (Optimized)' for all users by @AndyMik90 in 08aa2ff
+
+- fix: update default selected agent profile to 'auto' by @AndyMik90 in 37ace0a
+
+- style: enhance WorkspaceStatus component UI by @AndyMik90 in 3092155
+
+- fix: display correct merge target branch in worktree UI by @AndyMik90 in 2b96160
+
+- Improvement/refactor task sidebar to task modal by @AndyMik90 in 2a96f85
+
+- fix: extract human-readable title from spec.md when feature field is spec ID by @AndyMik90 in 8b59375
+
+- fix: task descriptions not showing for specs with compact markdown by @AndyMik90 in 7f12ef0
+
+- Add comprehensive theme system with Forest, Neo, Retro, Dusk, Ocean, and Lime color schemes by @AndyMik90 in ba776a3, e2b24e2, 7589046, e248256, 76c1bd7, bcbced2
+
+- Add ColorTheme type and configuration to app settings by @AndyMik90 in 2ca89ce, c505d6e, a75c0a9
+
+- Implement device code authentication flow with timeout handling and fallback URL display by @AndyMik90 in 5f26d39, 81e1536, 1a7cf40, 4a4ad6b, 6a4c1b4, b75a09c, e134c4c
+
+- fix(graphiti): address CodeRabbit review comments by @adryserage in 679b8cd
+
+- fix(lint): sort imports in Google provider files by @adryserage in 1a38a06
+
+## 2.6.0 - Multi-Provider Graphiti Support & Platform Fixes
+
+### ✨ New Features
+
+- **Google AI Provider for Graphiti**: Full Google AI (Gemini) support for both LLM and embeddings in the Memory Layer
+  - Add GoogleLLMClient with gemini-2.0-flash default model
+  - Add GoogleEmbedder with text-embedding-004 default model
+  - UI integration for Google API key configuration with link to Google AI Studio
+- **Ollama LLM Provider in UI**: Add Ollama as an LLM provider option in Graphiti onboarding wizard
+  - Ollama runs locally and doesn't require an API key
+  - Configure Base URL instead of API key for local inference
+- **LLM Provider Selection UI**: Add provider selection dropdown to Graphiti setup wizard for flexible backend configuration
+- **Per-Project GitHub Configuration**: UI clarity improvements for per-project GitHub org/repo settings
+
+### 🛠️ Improvements
+
+- Enhanced Graphiti provider factory to support Google AI alongside existing providers
+- Updated env-handlers to properly populate graphitiProviderConfig from .env files
+- Improved type definitions with proper Graphiti provider config properties in AppSettings
+- Better API key loading when switching between providers in settings
+
+### 🐛 Bug Fixes
+
+- **node-pty Migration**: Replaced node-pty with @lydell/node-pty for prebuilt Windows binaries
+  - Updated all imports to use @lydell/node-pty directly
+  - Fixed "Cannot find module 'node-pty'" startup error
+- **GitHub Organization Support**: Fixed repository support for GitHub organization accounts
+  - Add defensive array validation for GitHub issues API response
+- **Asyncio Deprecation**: Fixed asyncio deprecation warning by using get_running_loop() instead of get_event_loop()
+- Applied ruff formatting and fixed import sorting (I001) in Google provider files
+
+### 🔧 Other Changes
+
+- Added google-generativeai dependency to requirements.txt
+- Updated provider validation to include Google/Groq/HuggingFace type assertions
+
+---
+
+## What's Changed
+
+- fix(graphiti): address CodeRabbit review comments by @adryserage in 679b8cd
+- fix(lint): sort imports in Google provider files by @adryserage in 1a38a06
+- feat(graphiti): add Google AI as LLM and embedding provider by @adryserage in fe69106
+- fix: GitHub organization repository support by @mojaray2k in 873cafa
+- feat(ui): add LLM provider selection to Graphiti onboarding by @adryserage in 4750869
+- fix(types): add missing AppSettings properties for Graphiti providers by @adryserage in 6680ed4
+- feat(ui): add Ollama as LLM provider option for Graphiti by @adryserage in a3eee92
+- fix(ui): address PR review feedback for Graphiti provider selection by @adryserage in b8a419a
+- fix(deps): update imports to use @lydell/node-pty directly by @adryserage in 2b61ebb
+- fix(deps): replace node-pty with @lydell/node-pty for prebuilt binaries by @adryserage in e1aee6a
+- fix: add UI clarity for per-project GitHub configuration by @mojaray2k in c9745b6
+- fix: add defensive array validation for GitHub issues API response by @mojaray2k in b3636a5
+
+---
+
+## 2.5.5 - Enhanced Agent Reliability & Build Workflow
+
+### ✨ New Features
+
+- Required GitHub setup flow after Auto Claude initialization to ensure proper configuration
+- Atomic log saving mechanism to prevent log file corruption during concurrent operations
+- Per-session model and thinking level selection in insights management
+- Multi-auth token support and ANTHROPIC_BASE_URL passthrough for flexible authentication
+- Comprehensive DEBUG logging at Claude SDK invocation points for improved troubleshooting
+- Auto-download of prebuilt node-pty binaries for Windows environments
+- Enhanced merge workflow with current branch detection for accurate change previews
+- Phase configuration module and enhanced agent profiles for improved flexibility
+- Stage-only merge handling with comprehensive verification checks
+- Authentication failure detection system with patterns and validation checks across agent pipeline
+
+### 🛠️ Improvements
+
+- Changed default agent profile from 'balanced' to 'auto' for more adaptive behavior
+- Better GitHub issue tracking and improved user experience in issue management
+- Improved merge preview accuracy using git diff counts for file statistics
+- Preserved roadmap generation state when switching between projects
+- Enhanced agent profiles with phase configuration support
+
+### 🐛 Bug Fixes
+
+- Resolved CI test failures and improved merge preview reliability
+- Fixed CI failures related to linting, formatting, and tests
+- Prevented dialog skip during project initialization flow
+- Updated model IDs for Sonnet and Haiku to match current Claude versions
+- Fixed branch namespace conflict detection to prevent worktree creation failures
+- Removed duplicate LINEAR_API_KEY checks and consolidated imports
+- Python 3.10+ version requirement enforced with proper version checking
+- Prevented command injection vulnerabilities in GitHub API calls
+
+### 🔧 Other Changes
+
+- Code cleanup and test fixture updates
+- Removed redundant auto-claude/specs directory structure
+- Untracked .auto-claude directory to respect gitignore rules
+
+---
+
+## What's Changed
+
+- fix: resolve CI test failures and improve merge preview by @AndyMik90 in de2eccd
+- chore: code cleanup and test fixture updates by @AndyMik90 in 948db57
+- refactor: change default agent profile from 'balanced' to 'auto' by @AndyMik90 in f98a13e
+- security: prevent command injection in GitHub API calls by @AndyMik90 in 24ff491
+- fix: resolve CI failures (lint, format, test) by @AndyMik90 in a8f2d0b
+- fix: use git diff count for totalFiles in merge preview by @AndyMik90 in 46d2536
+- feat: enhance stage-only merge handling with verification checks by @AndyMik90 in 7153558
+- feat: introduce phase configuration module and enhance agent profiles by @AndyMik90 in 2672528
+- fix: preserve roadmap generation state when switching projects by @AndyMik90 in 569e921
+- feat: add required GitHub setup flow after Auto Claude initialization by @AndyMik90 in 03ccce5
+- chore: remove redundant auto-claude/specs directory by @AndyMik90 in 64d5170
+- chore: untrack .auto-claude directory (should be gitignored) by @AndyMik90 in 0710c13
+- fix: prevent dialog skip during project initialization by @AndyMik90 in 56cedec
+- feat: enhance merge workflow by detecting current branch by @AndyMik90 in c0c8067
+- fix: update model IDs for Sonnet and Haiku by @AndyMik90 in 059315d
+- feat: add comprehensive DEBUG logging and fix lint errors by @AndyMik90 in 99cf21e
+- feat: implement atomic log saving to prevent corruption by @AndyMik90 in da5e26b
+- feat: add better github issue tracking and UX by @AndyMik90 in c957eaa
+- feat: add comprehensive DEBUG logging to Claude SDK invocation points by @AndyMik90 in 73d01c0
+- feat: auto-download prebuilt node-pty binaries for Windows by @AndyMik90 in 41a507f
+- feat(insights): add per-session model and thinking level selection by @AndyMik90 in e02aa59
+- fix: require Python 3.10+ and add version check by @AndyMik90 in 9a5ca8c
+- fix: detect branch namespace conflict blocking worktree creation by @AndyMik90 in 63a1d3c
+- fix: remove duplicate LINEAR_API_KEY check and consolidate imports by @Jacob in 7d351e3
+- feat: add multi-auth token support and ANTHROPIC_BASE_URL passthrough by @Jacob in 9dea155
+
 ## 2.5.0 - Roadmap Intelligence & Workflow Refinements
 
 ### ✨ New Features
@@ -178,7 +621,7 @@
 
 - Restructured SortableFeatureCard badge layout for improved visual presentation
 
-Bug Fixes: 
+Bug Fixes:
 - Fixed spec runner path configuration for more reliable task execution
 
 ---
@@ -481,4 +924,4 @@ Bug Fixes:
 - **Draft Auto-Save**: Task creation state is now automatically saved when you navigate away, preventing accidental loss of work-in-progress.
 
 ### Bug Fixes
-- Fixed task editing to support the same comprehensive options available in new task creation
+- Fixed task editing to support the same comprehensive options available in new task creation

CLAUDE.md

@@ -81,6 +81,21 @@ auto-claude/.venv/bin/pytest tests/ -m "not slow"
 python auto-claude/validate_spec.py --spec-dir auto-claude/specs/001-feature --checkpoint all
 ```
 
+### Releases
+```bash
+# Automated version bump and release (recommended)
+node scripts/bump-version.js patch   # 2.5.5 -> 2.5.6
+node scripts/bump-version.js minor   # 2.5.5 -> 2.6.0
+node scripts/bump-version.js major   # 2.5.5 -> 3.0.0
+node scripts/bump-version.js 2.6.0   # Set specific version
+
+# Then push to trigger GitHub release workflows
+git push origin main
+git push origin v2.6.0
+```
+
+See [RELEASE.md](RELEASE.md) for detailed release process documentation.
+
 ## Architecture
 
 ### Core Pipeline
@@ -103,7 +118,7 @@ python auto-claude/validate_spec.py --spec-dir auto-claude/specs/001-feature --c
 - **worktree.py** - Git worktree isolation for safe feature development
 - **memory.py** - File-based session memory (primary, always-available storage)
 - **graphiti_memory.py** - Optional graph-based cross-session memory with semantic search
-- **graphiti_providers.py** - Multi-provider factory for Graphiti (OpenAI, Anthropic, Azure, Ollama)
+- **graphiti_providers.py** - Multi-provider factory for Graphiti (OpenAI, Anthropic, Azure, Ollama, Google AI)
 - **graphiti_config.py** - Configuration and validation for Graphiti integration
 - **linear_updater.py** - Optional Linear integration for progress tracking
 
@@ -174,11 +189,17 @@ Dual-layer memory architecture:
 - Session insights, patterns, gotchas, codebase map
 
 **Graphiti Memory (Optional Enhancement)** - `graphiti_memory.py`
-- Graph database with semantic search (FalkorDB)
+- Graph database with semantic search (LadybugDB - embedded, no Docker)
 - Cross-session context retrieval
-- Multi-provider support (V2):
-  - LLM: OpenAI, Anthropic, Azure OpenAI, Ollama
-  - Embedders: OpenAI, Voyage AI, Azure OpenAI, Ollama
+- Requires Python 3.12+
+- Multi-provider support:
+  - LLM: OpenAI, Anthropic, Azure OpenAI, Ollama, Google AI (Gemini)
+  - Embedders: OpenAI, Voyage AI, Azure OpenAI, Ollama, Google AI
+
+```bash
+# Setup (requires Python 3.12+)
+pip install real_ladybug graphiti-core
+```
 
 Enable with: `GRAPHITI_ENABLED=true` + provider credentials. See `.env.example`.
 

CONTRIBUTING.md

@@ -8,6 +8,7 @@ Thank you for your interest in contributing to Auto Claude! This document provid
 - [Development Setup](#development-setup)
   - [Python Backend](#python-backend)
   - [Electron Frontend](#electron-frontend)
+- [Running from Source](#running-from-source)
 - [Pre-commit Hooks](#pre-commit-hooks)
 - [Code Style](#code-style)
 - [Testing](#testing)
@@ -28,7 +29,6 @@ Before contributing, ensure you have the following installed:
 - **pnpm** - Package manager for the frontend (`npm install -g pnpm`)
 - **uv** (recommended) or **pip** - Python package manager
 - **Git** - Version control
-- **Docker** (optional) - For running FalkorDB if using Graphiti memory
 
 ## Development Setup
 
@@ -80,6 +80,58 @@ pnpm build
 pnpm package
 ```
 
+## Running from Source
+
+If you want to run Auto Claude from source (for development or testing unreleased features), follow these steps:
+
+### Step 1: Clone and Set Up Python Backend
+
+```bash
+git clone https://github.com/AndyMik90/Auto-Claude.git
+cd Auto-Claude/auto-claude
+
+# Using uv (recommended)
+uv venv && uv pip install -r requirements.txt
+
+# Or using standard Python
+python3 -m venv .venv
+source .venv/bin/activate  # On Windows: .venv\Scripts\activate
+pip install -r requirements.txt
+
+# Set up environment
+cp .env.example .env
+# Edit .env and add your CLAUDE_CODE_OAUTH_TOKEN (get it via: claude setup-token)
+```
+
+### Step 2: Run the Desktop UI
+
+```bash
+cd ../auto-claude-ui
+
+# Install dependencies
+pnpm install
+
+# Development mode (hot reload)
+pnpm dev
+
+# Or production build
+pnpm run build && pnpm run start
+```
+
+<details>
+<summary><b>Windows users:</b> If installation fails with node-gyp errors, click here</summary>
+
+Auto Claude automatically downloads prebuilt binaries for Windows. If prebuilts aren't available for your Electron version yet, you'll need Visual Studio Build Tools:
+
+1. Download [Visual Studio Build Tools 2022](https://visualstudio.microsoft.com/visual-cpp-build-tools/)
+2. Select "Desktop development with C++" workload
+3. In "Individual Components", add "MSVC v143 - VS 2022 C++ x64/x86 Spectre-mitigated libs"
+4. Restart terminal and run `pnpm install` again
+
+</details>
+
+> **Note:** For regular usage, we recommend downloading the pre-built releases from [GitHub Releases](https://github.com/AndyMik90/Auto-Claude/releases). Running from source is primarily for contributors and those testing unreleased features.
+
 ## Pre-commit Hooks
 
 We use [pre-commit](https://pre-commit.com/) to run linting and formatting checks before each commit. This ensures code quality and consistency across the project.

INVESTIGATION.md

@@ -0,0 +1,641 @@
+# Investigation: v2.7.1 Release Artifacts Issue
+
+## Investigation Date
+
+2025-12-25
+
+## Summary
+
+The v2.7.1 release has **incorrect files attached**. All artifacts have v2.7.0 in their filenames, indicating the wrong build artifacts were uploaded.
+
+---
+
+## Phase 1: Reproduce and Verify Issue
+
+### Subtask 1-1: Current v2.7.1 Assets
+
+**Command:** `gh release view v2.7.1 --json assets -q '.assets[].name'`
+
+**Release Metadata:**
+- Tag Name: v2.7.1
+- Release Name: v2.7.1
+- Published At: 2025-12-22T13:35:38Z
+- Is Draft: false
+- Is Prerelease: false
+
+**Files Currently Attached to v2.7.1:**
+
+| File Name | Size (bytes) | Expected Name |
+|-----------|-------------|---------------|
+| Auto-Claude-2.7.0-darwin-arm64.dmg | 124,187,073 | Auto-Claude-2.7.1-darwin-arm64.dmg |
+| Auto-Claude-2.7.0-darwin-arm64.zip | 117,694,085 | Auto-Claude-2.7.1-darwin-arm64.zip |
+| Auto-Claude-2.7.0-darwin-x64.dmg | 130,635,398 | Auto-Claude-2.7.1-darwin-x64.dmg |
+| Auto-Claude-2.7.0-darwin-x64.zip | 124,176,354 | Auto-Claude-2.7.1-darwin-x64.zip |
+| Auto-Claude-2.7.0-linux-amd64.deb | 104,558,694 | Auto-Claude-2.7.1-linux-amd64.deb |
+| Auto-Claude-2.7.0-linux-x86_64.AppImage | 145,482,885 | Auto-Claude-2.7.1-linux-x86_64.AppImage |
+| Auto-Claude-2.7.0-win32-x64.exe | 101,941,972 | Auto-Claude-2.7.1-win32-x64.exe |
+| checksums.sha256 | 718 | checksums.sha256 (with v2.7.1 filenames) |
+
+### Issue Confirmed
+
+**Problem:** All 7 platform artifacts attached to v2.7.1 have "2.7.0" in their filename instead of "2.7.1".
+
+**Impact:**
+- Users downloading v2.7.1 are receiving v2.7.0 binaries
+- File naming does not match the release version
+- Checksums file likely references v2.7.0 filenames
+- Auto-update mechanisms may be confused by version mismatch
+
+**Evidence:**
+```
+Files attached to v2.7.1:
+- Auto-Claude-2.7.0-darwin-arm64.dmg   (WRONG - should be 2.7.1)
+- Auto-Claude-2.7.0-darwin-arm64.zip   (WRONG - should be 2.7.1)
+- Auto-Claude-2.7.0-darwin-x64.dmg     (WRONG - should be 2.7.1)
+- Auto-Claude-2.7.0-darwin-x64.zip     (WRONG - should be 2.7.1)
+- Auto-Claude-2.7.0-linux-amd64.deb    (WRONG - should be 2.7.1)
+- Auto-Claude-2.7.0-linux-x86_64.AppImage (WRONG - should be 2.7.1)
+- Auto-Claude-2.7.0-win32-x64.exe      (WRONG - should be 2.7.1)
+- checksums.sha256                      (likely references wrong filenames)
+```
+
+---
+
+### Subtask 1-2: Comparison with v2.7.0 and Expected Naming
+
+**Command:** `gh release view v2.7.0 --json assets -q '.assets[].name'`
+
+#### v2.7.0 Release Analysis
+
+**Release Metadata:**
+- Tag Name: v2.7.0
+- Release Name: v2.7.0
+- Published At: 2025-12-22T13:19:13Z
+- Target Commitish: main
+- Is Draft: false
+- Is Prerelease: false
+
+**Critical Finding:** v2.7.0 has **NO assets attached** (empty assets array).
+
+#### Release Timeline
+
+| Release | Published At | Assets Count | Status |
+|---------|-------------|--------------|--------|
+| v2.7.0  | 2025-12-22T13:19:13Z | 0 | No files attached |
+| v2.7.1  | 2025-12-22T13:35:38Z | 8 | Wrong version in filenames |
+| v2.7.2  | 2025-12-22T13:52:51Z | ? | Draft release |
+
+**Observation:** v2.7.0 was published 16 minutes before v2.7.1, but has no artifacts attached.
+
+#### Checksums File Analysis
+
+The `checksums.sha256` file attached to v2.7.1 contains:
+```
+0a0094ff3e52609665f6f0d6d54180dbfc592956f91ef2cdd94e43a61b6b24d2  ./Auto-Claude-2.7.0-darwin-arm64.dmg
+43b168f3073d60644bb111c8fa548369431dc448e67700ed526cb4cad61034e0  ./Auto-Claude-2.7.0-darwin-arm64.zip
+5150cbba934fbeb3d97309a493cc8ef3c035e9ec38b31f01382d628025f5c451  ./Auto-Claude-2.7.0-darwin-x64.dmg
+ea9139277290a8189f799d00bc3cd1aaf81a16e890ff90327eca01a4cce73e61  ./Auto-Claude-2.7.0-darwin-x64.zip
+078b2ba6a2594bf048932776dc31a45e59cd9cb23b34b2cf2f810f4101f04736  ./Auto-Claude-2.7.0-linux-amd64.deb
+1feb6b9be348a5e23238e009dbc1ce8b2788103a262cd856613332b3ab1711e9  ./Auto-Claude-2.7.0-linux-x86_64.AppImage
+25383314b3bc032ceaf8a8416d5383879ed351c906f03175b8533047647a612d  ./Auto-Claude-2.7.0-win32-x64.exe
+```
+
+**Issue:** Checksums file also references v2.7.0 filenames, confirming the build was run with v2.7.0 version.
+
+#### Expected Naming Pattern (from release.yml)
+
+Based on the release workflow analysis, artifacts follow this naming convention:
+```
+Auto-Claude-{version}-{platform}-{arch}.{ext}
+```
+
+Where version comes from `package.json` in `auto-claude-ui/`.
+
+**Expected v2.7.1 Artifacts:**
+| Expected Filename | Actual Filename (Wrong) |
+|-------------------|-------------------------|
+| Auto-Claude-2.7.1-darwin-arm64.dmg | Auto-Claude-2.7.0-darwin-arm64.dmg |
+| Auto-Claude-2.7.1-darwin-arm64.zip | Auto-Claude-2.7.0-darwin-arm64.zip |
+| Auto-Claude-2.7.1-darwin-x64.dmg | Auto-Claude-2.7.0-darwin-x64.dmg |
+| Auto-Claude-2.7.1-darwin-x64.zip | Auto-Claude-2.7.0-darwin-x64.zip |
+| Auto-Claude-2.7.1-linux-amd64.deb | Auto-Claude-2.7.0-linux-amd64.deb |
+| Auto-Claude-2.7.1-linux-x86_64.AppImage | Auto-Claude-2.7.0-linux-x86_64.AppImage |
+| Auto-Claude-2.7.1-win32-x64.exe | Auto-Claude-2.7.0-win32-x64.exe |
+| checksums.sha256 (v2.7.1 refs) | checksums.sha256 (v2.7.0 refs) |
+
+#### Hypothesis
+
+The evidence suggests one of the following scenarios:
+
+1. **Tag/Version Mismatch:** The v2.7.1 tag may point to a commit where `package.json` still had version `2.7.0`
+2. **Workflow Re-run:** The v2.7.1 release may have been created by re-running the v2.7.0 workflow artifacts
+3. **Manual Upload Error:** Artifacts from v2.7.0 were manually attached to the v2.7.1 release
+4. **Artifact Caching:** Old workflow artifacts were incorrectly reused for v2.7.1
+
+**Next step:** Check git tags and package.json versions to determine root cause.
+
+---
+
+### Subtask 1-3: Package.json Version and Git State Analysis
+
+**Commands Used:**
+- `git show v2.7.1:auto-claude-ui/package.json | jq -r '.version'`
+- `git show v2.7.0:auto-claude-ui/package.json | jq -r '.version'`
+- `git log --oneline v2.7.0..v2.7.1`
+- `git rev-parse v2.7.1^{commit}`
+
+#### Current Package.json State
+
+| Location | Current Version |
+|----------|-----------------|
+| `auto-claude-ui/package.json` (HEAD) | 2.7.1 |
+
+**Note:** The subtask referenced `apps/frontend/package.json`, but the actual path is `auto-claude-ui/package.json`.
+
+#### Version at Git Tags
+
+| Tag | Commit | package.json Version | Expected |
+|-----|--------|---------------------|----------|
+| v2.7.0 | `fe7290a8` | 2.6.5 | 2.7.0 |
+| v2.7.1 | `772a5006` | **2.7.0** ❌ | 2.7.1 |
+
+#### Commit Timeline
+
+```
+fc2075dd auto-claude: subtask-1-2 - Compare v2.7.1 artifacts...
+ff033a8e auto-claude: subtask-1-1 - List all files...
+8db71f3d Update version to 2.7.1 in package.json    <-- Version bump (AFTER tag)
+772a5006 2.7.1                                      <-- v2.7.1 TAG placed here
+d23fcd86 Enhance VirusTotal scan error handling...
+...more commits...
+fe7290a8 Release v2.7.0...                          <-- v2.7.0 TAG placed here
+```
+
+#### Root Cause Identified ✅
+
+**Problem:** The `v2.7.1` tag was placed on commit `772a5006` BEFORE the `package.json` version was updated to `2.7.1`.
+
+**Timeline of error:**
+1. Commit `772a5006` created with message "2.7.1" - tag `v2.7.1` placed here
+2. At this commit, `package.json` still contained version `2.7.0`
+3. The release workflow triggered on tag push, building with version `2.7.0` from `package.json`
+4. All artifacts named with `2.7.0` because that's what was in `package.json`
+5. Commit `8db71f3d` later updated `package.json` to `2.7.1` (but tag was already pushed)
+
+**This is a "tag before version bump" error.**
+
+The release workflow correctly read the version from `package.json`, but the tag was created before the version was bumped. The naming convention `${productName}-${version}-${platform}-${arch}.${ext}` correctly used version `2.7.0` because that's what was in `package.json` at the tagged commit.
+
+#### Verification of Build Configuration
+
+From `auto-claude-ui/package.json`:
+```json
+"build": {
+  "artifactName": "${productName}-${version}-${platform}-${arch}.${ext}",
+  ...
+}
+```
+
+This confirms the version is sourced from `package.json` during the build process.
+
+#### Git State Summary
+
+| Metric | Value |
+|--------|-------|
+| Current Branch | `auto-claude/009-latest-release-v2-7-1-has-wrong-files-attached` |
+| Working Tree | Clean |
+| Current HEAD package.json | 2.7.1 |
+| v2.7.1 tag package.json | 2.7.0 ❌ |
+| v2.7.0 tag package.json | 2.6.5 ❌ |
+
+**Note:** Both v2.7.0 and v2.7.1 tags have version mismatches in `package.json`, indicating a pattern of tagging before version bumping.
+
+---
+
+## Root Cause Summary
+
+| Factor | Finding |
+|--------|---------|
+| What happened? | v2.7.1 tag placed before package.json version bump |
+| Why? | Incorrect release process: tag first, version bump second |
+| Impact | All 7 artifacts have v2.7.0 in filename |
+| Evidence | `git show v2.7.1:auto-claude-ui/package.json` shows version 2.7.0 |
+
+---
+
+## Phase 2: Root Cause Analysis
+
+### Subtask 2-1: Inspect v2.7.1 Git Tag and Commit
+
+**Commands Used:**
+```bash
+git log -1 v2.7.1 --format='%H %s %ci'
+git show v2.7.1 --format='Commit: %H%nAuthor: %an <%ae>%nDate: %ci%nMessage: %s' --no-patch
+git tag -l v2.7.1 -n1
+git cat-file -t v2.7.1
+git show v2.7.1:auto-claude-ui/package.json | head -10 | grep version
+```
+
+#### Tag Details
+
+| Property | Value |
+|----------|-------|
+| Tag Name | v2.7.1 |
+| Tag Type | Lightweight (commit reference, not annotated) |
+| Points To | `772a5006d45487b600ce4079bae1c98f9ccf6b2e` |
+
+#### Tagged Commit Details
+
+| Property | Value |
+|----------|-------|
+| Commit Hash | `772a5006d45487b600ce4079bae1c98f9ccf6b2e` |
+| Author | AndyMik90 <andre@mikalsenutvikling.no> |
+| Commit Date | 2025-12-22 14:35:30 +0100 |
+| Commit Message | `2.7.1` |
+| package.json Version | **2.7.0** (MISMATCH) |
+
+#### Verification Output
+
+```
+$ git log -1 v2.7.1 --format='%H %s %ci'
+772a5006d45487b600ce4079bae1c98f9ccf6b2e 2.7.1 2025-12-22 14:35:30 +0100
+
+$ git show v2.7.1:auto-claude-ui/package.json | grep version
+  "version": "2.7.0",
+```
+
+#### Commit Context
+
+```
+$ git log -3 --oneline v2.7.1
+772a5006 2.7.1                                        <-- v2.7.1 TAG HERE
+d23fcd86 Enhance VirusTotal scan error handling...
+326118bd Refactor macOS build workflow...
+```
+
+#### Analysis
+
+1. **Tag Type:** The tag is a lightweight tag (just a commit reference), not an annotated tag. This means there's no separate tag object with metadata, author, or message.
+
+2. **Commit Message vs Version:** The commit message says "2.7.1" but the `package.json` at this commit still contains version `2.7.0`. This is the source of the mismatch.
+
+3. **Release Workflow Behavior:** When the GitHub release workflow triggered on tag push `v2.7.1`:
+   - It checked out commit `772a5006`
+   - It read version from `auto-claude-ui/package.json` which was `2.7.0`
+   - It built artifacts with `2.7.0` in the filename
+   - It uploaded these incorrectly-versioned artifacts to the v2.7.1 release
+
+4. **Timeline Confirmation:**
+   - Tag created: 2025-12-22 14:35:30 +0100
+   - Release published: 2025-12-22T13:35:38Z (same time, UTC)
+   - Version bump commit `8db71f3d` happened AFTER this
+
+#### Root Cause Confirmed
+
+The v2.7.1 tag points to a commit where `package.json` still had version `2.7.0`. This is a **"tag before version bump"** error in the release process.
+
+The correct sequence should have been:
+1. First: Bump package.json version to 2.7.1
+2. Second: Commit the version bump
+3. Third: Create and push the v2.7.1 tag
+
+What actually happened:
+1. Created tag v2.7.1 on commit with package.json version 2.7.0
+2. Workflow triggered and built with wrong version
+3. Version bump to 2.7.1 committed afterwards (too late)
+
+---
+
+### Subtask 2-2: GitHub Actions Workflow Run Analysis
+
+**Commands Used:**
+```bash
+gh run list --workflow=release.yml --limit=20
+gh run view 20433472030 --json conclusion,status,headSha,event,jobs
+gh run view 20433472034 --json conclusion,status,headSha,jobs  # Validate Version workflow
+```
+
+#### Release Workflow Run Details
+
+| Property | Value |
+|----------|-------|
+| Run ID | 20433472030 |
+| Status | Completed |
+| Conclusion | **success** |
+| Event | push (tag v2.7.1) |
+| Head SHA | `772a5006d45487b600ce4079bae1c98f9ccf6b2e` |
+| Created At | 2025-12-22T13:35:39Z |
+| Updated At | 2025-12-22T13:53:02Z |
+| Total Duration | ~17 minutes |
+
+#### Build Jobs Summary
+
+| Job | Status | Duration | Started | Completed |
+|-----|--------|----------|---------|-----------|
+| build-linux | ✅ success | ~2.5 min | 13:35:42Z | 13:38:15Z |
+| build-windows | ✅ success | ~4.5 min | 13:35:41Z | 13:40:11Z |
+| build-macos-intel | ✅ success | ~7 min | 13:35:42Z | 13:42:52Z |
+| build-macos-arm64 | ✅ success | ~5.5 min | 13:35:42Z | 13:41:12Z |
+| create-release | ✅ success | ~10 min | 13:42:55Z | 13:53:01Z |
+
+#### Create-Release Job Steps
+
+| Step | Conclusion |
+|------|------------|
+| Set up job | ✅ success |
+| Run actions/checkout@v4 | ✅ success |
+| Download all artifacts | ✅ success |
+| Flatten and validate artifacts | ✅ success |
+| Generate checksums | ✅ success |
+| Scan with VirusTotal | ✅ success |
+| Dry run summary | ⏭️ skipped |
+| Generate changelog | ✅ success |
+| Create Release | ✅ success |
+
+**Workflow Analysis:** The release workflow executed **successfully** - all build jobs and the release creation completed without errors. The workflow correctly:
+1. Checked out commit `772a5006d45487b600ce4079bae1c98f9ccf6b2e`
+2. Built artifacts for all platforms (Linux, Windows, macOS Intel, macOS ARM64)
+3. Generated checksums
+4. Ran VirusTotal scans
+5. Created the GitHub release with artifacts
+
+**Key Finding:** The workflow operated as designed. The problem was the **input** (source code at tagged commit), not the **workflow logic**.
+
+---
+
+#### 🚨 CRITICAL: Validate Version Workflow FAILED
+
+**Run ID:** 20433472034
+**Conclusion:** ❌ **FAILURE**
+
+| Step | Conclusion |
+|------|------------|
+| Set up job | ✅ success |
+| Checkout | ✅ success |
+| Extract version from tag | ✅ success |
+| Extract version from package.json | ✅ success |
+| **Compare versions** | ❌ **FAILURE** |
+| Version validation result | ⏭️ skipped |
+
+**What Happened:**
+1. The `validate-version.yml` workflow triggered on the v2.7.1 tag push
+2. It correctly detected that:
+   - Tag version: `2.7.1`
+   - package.json version: `2.7.0`
+3. It **FAILED** with an error because versions didn't match
+
+**Why Didn't This Stop the Release?**
+
+The `validate-version.yml` and `release.yml` workflows are **independent**:
+- Both trigger on `push: tags: - 'v*'`
+- They run in **parallel**, not sequentially
+- The release workflow has **no dependency** on the validation workflow
+- Even though validation failed, the release proceeded and succeeded
+
+**Validation Workflow (from `.github/workflows/validate-version.yml`):**
+```yaml
+on:
+  push:
+    tags:
+      - 'v*'
+```
+
+**Expected Output from Failed Validation:**
+```
+❌ ERROR: Version mismatch detected!
+
+The version in package.json (2.7.0) does not match
+the git tag version (2.7.1).
+
+To fix this:
+  1. Delete this tag: git tag -d v2.7.1
+  2. Update package.json version to 2.7.1
+  3. Commit the change
+  4. Recreate the tag: git tag -a v2.7.1 -m 'Release v2.7.1'
+```
+
+---
+
+#### Workflow Architecture Issue
+
+```
+Tag Push (v2.7.1)
+       │
+       ├──────────────────────────────────┐
+       │                                  │
+       ▼                                  ▼
+┌──────────────────┐            ┌──────────────────┐
+│  validate-version │            │     release      │
+│    ❌ FAILED      │            │   ✅ SUCCESS     │
+│ (detected error)  │  ← No    │  (built wrong     │
+│                   │   link →  │   version)        │
+└──────────────────┘            └──────────────────┘
+```
+
+**Problem:** The validation workflow runs but cannot **block** the release workflow.
+
+---
+
+#### Other v2.7.1 Workflows
+
+| Workflow | Run ID | Conclusion | Notes |
+|----------|--------|------------|-------|
+| Release | 20433472030 | ✅ success | Built and released v2.7.0 files |
+| Validate Version | 20433472034 | ❌ failure | Detected mismatch but couldn't stop release |
+| Discord Release Notification | 20433472029 | ✅ success | Notified Discord about "v2.7.1" |
+| Test on Tag | 20433472046 | ✅ success | Tests passed |
+| Build Native Module Prebuilds | 20433472017 | ❌ failure | Separate prebuilt module issue |
+
+---
+
+#### Root Cause Confirmation
+
+The GitHub Actions analysis confirms:
+
+1. **The release workflow worked correctly** - it built exactly what was in the source code at the tagged commit
+2. **The validation workflow detected the problem** - it correctly identified the version mismatch
+3. **The workflows are not connected** - validation failure could not prevent the release
+4. **The artifacts are from the right commit but wrong version** - v2.7.1 release contains v2.7.0 artifacts because that's what package.json said at commit `772a5006`
+
+---
+
+### Subtask 2-3: Root Cause Statement and Fix Options
+
+#### Root Cause Statement
+
+**Summary:** The v2.7.1 release contains v2.7.0 artifacts because the git tag was created BEFORE the `package.json` version was updated.
+
+**Root Cause:** "Tag Before Version Bump" Error
+
+| Factor | Details |
+|--------|---------|
+| **What happened** | The `v2.7.1` git tag was placed on commit `772a5006` which still had `package.json` version `2.7.0` |
+| **Why it happened** | Incorrect release procedure: tag was created before version bump was committed |
+| **How it propagated** | The release workflow correctly built from the tagged commit, reading version `2.7.0` from `package.json` |
+| **Why it wasn't caught** | The `validate-version.yml` workflow detected the mismatch but runs in parallel with `release.yml` and cannot block it |
+
+**Evidence Chain:**
+1. `git show v2.7.1:auto-claude-ui/package.json` → version `2.7.0`
+2. Release workflow run #20433472030 → built from commit `772a5006` → success
+3. Validate-version workflow run #20433472034 → detected mismatch → **FAILED** (but couldn't stop release)
+4. All 7 artifacts named `Auto-Claude-2.7.0-*` instead of `Auto-Claude-2.7.1-*`
+
+**Contributing Factors:**
+- Lightweight tag (no metadata) - easier to create on wrong commit
+- No pre-tag validation hook or script
+- Independent parallel workflows with no blocking dependency
+- Version in `package.json` is the single source of truth for artifact naming
+
+---
+
+#### Fix Options
+
+##### Option A: Recreate v2.7.1 Tag and Release (RECOMMENDED)
+
+**Description:** Delete the current v2.7.1 tag and release, then create a new tag on a commit where `package.json` has version `2.7.1`.
+
+**Steps:**
+1. Delete the v2.7.1 GitHub release: `gh release delete v2.7.1 --cleanup-tag --yes`
+2. Identify correct commit: `git log --oneline | grep -A1 "Update version to 2.7.1"`
+3. Create new tag at correct commit: `git tag v2.7.1 8db71f3d && git push origin v2.7.1`
+4. Release workflow will trigger automatically with correct version
+5. Verify new artifacts have `2.7.1` in filenames
+
+**Pros:**
+- ✅ Users get the correct version they expect (v2.7.1)
+- ✅ Maintains clean version history
+- ✅ Checksums will match the correct filenames
+- ✅ Auto-update mechanisms will work correctly
+- ✅ No need to update documentation or links
+
+**Cons:**
+- ⚠️ Users who already downloaded v2.7.1 (v2.7.0 files) may be confused
+- ⚠️ Requires deleting and recreating the release
+- ⚠️ Brief window where v2.7.1 doesn't exist
+
+**Risk Level:** Medium - temporary unavailability, but correct outcome
+
+---
+
+##### Option B: Publish v2.7.2 with Correct Files
+
+**Description:** Leave v2.7.1 as-is (deprecated) and publish v2.7.2 with the correct build.
+
+**Steps:**
+1. Mark v2.7.1 as deprecated in release notes
+2. Bump package.json to 2.7.2
+3. Create and push v2.7.2 tag
+4. Publish v2.7.2 release with correct artifacts
+5. Update download links/documentation to point to v2.7.2
+
+**Pros:**
+- ✅ No disruption to existing v2.7.1 downloads
+- ✅ Preserves release history for audit trail
+- ✅ Clear indication that v2.7.2 supersedes v2.7.1
+
+**Cons:**
+- ❌ Version number gap in release history (no "real" 2.7.1)
+- ❌ Confusing version progression for users
+- ❌ Requires updating all download links and documentation
+- ❌ Users may still download deprecated v2.7.1
+
+**Risk Level:** Low - no deletion, but messy version history
+
+---
+
+##### Option C: Manual File Upload with --clobber
+
+**Description:** Build correct v2.7.1 artifacts locally and upload to replace existing files.
+
+**Steps:**
+1. Checkout commit with package.json version 2.7.1
+2. Build all platform artifacts locally (or trigger workflow to download)
+3. Delete existing assets: `gh release delete-asset v2.7.1 Auto-Claude-2.7.0-* --yes`
+4. Upload correct files: `gh release upload v2.7.1 ./dist/* --clobber`
+5. Update checksums file
+
+**Pros:**
+- ✅ Keeps same release and tag
+- ✅ No temporary unavailability window
+
+**Cons:**
+- ❌ Complex manual process prone to errors
+- ❌ Requires local build environment for all platforms (macOS, Windows, Linux)
+- ❌ May not match workflow-built artifacts exactly
+- ❌ Code signing could be different than CI
+- ❌ Does not address underlying tag/commit mismatch
+
+**Risk Level:** High - manual process, potential signing issues
+
+---
+
+#### Recommendation
+
+**Recommended Option: A - Recreate v2.7.1 Tag and Release**
+
+**Rationale:**
+1. **Correctness**: The tag should point to a commit where the codebase reflects v2.7.1
+2. **Automation**: Letting the release workflow rebuild ensures identical CI artifacts
+3. **User Experience**: Users expect v2.7.1 to contain 2.7.1 code and files
+4. **Maintainability**: Clean version history is easier to manage long-term
+5. **Auto-updates**: Electron auto-updater relies on version matching
+
+**Implementation Priority:**
+1. First: Fix v2.7.1 release (Option A)
+2. Then: Update workflow to prevent recurrence (make validation blocking)
+
+---
+
+#### Process Improvements Required
+
+Regardless of fix option chosen, these changes should be implemented to prevent recurrence:
+
+| Improvement | Priority | Description |
+|-------------|----------|-------------|
+| Make validation blocking | HIGH | Modify `release.yml` to depend on `validate-version.yml` passing |
+| Add pre-tag script | MEDIUM | Create script that validates version before allowing tag creation |
+| Use annotated tags | LOW | Annotated tags include metadata and require explicit message |
+| Document release procedure | MEDIUM | Create runbook for correct release process |
+
+**Workflow Architecture Change:**
+```yaml
+# Before (parallel, independent):
+Tag Push → release.yml (runs)
+         → validate-version.yml (runs, but can't block)
+
+# After (sequential, dependent):
+Tag Push → validate-version.yml (runs first)
+           ↓ success required
+         → release.yml (only runs if validation passes)
+```
+
+---
+
+## Next Steps
+
+1. ~~**Subtask 1-1:** Verify v2.7.1 assets~~ ✅ Complete
+2. ~~**Subtask 1-2:** Compare with v2.7.0 release and verify expected naming pattern~~ ✅ Complete
+3. ~~**Subtask 1-3:** Check package.json version and git state~~ ✅ Complete - ROOT CAUSE IDENTIFIED
+4. ~~**Subtask 2-1:** Inspect v2.7.1 git tag and commit~~ ✅ Complete - TAG/COMMIT MISMATCH CONFIRMED
+5. ~~**Subtask 2-2:** Check release workflow runs~~ ✅ Complete - VALIDATION DETECTED BUT COULDN'T STOP RELEASE
+6. ~~**Subtask 2-3:** Document fix options~~ ✅ Complete - 3 OPTIONS DOCUMENTED, OPTION A RECOMMENDED
+7. **Phase 3:** Implement fix (re-upload correct files or publish v2.7.2)
+8. **Phase 4:** Add validation to prevent future occurrences
+
+---
+
+## Status: Phase 2 Complete ✅
+
+**Root Cause:** The v2.7.1 tag was created on commit `772a5006` which still had `package.json` version `2.7.0`. The validation workflow detected this but couldn't stop the release workflow.
+
+**Key Findings from Workflow Analysis:**
+- Release workflow (ID: 20433472030): ✅ Success - correctly built from tagged commit
+- Validate Version workflow (ID: 20433472034): ❌ Failed - correctly detected version mismatch
+- The workflows run in parallel with no dependency relationship
+
+**Recommended Fix:** Option A - Delete v2.7.1 tag and release, recreate tag at commit `8db71f3d` where `package.json` has version `2.7.1`, let workflow rebuild.
+
+**Process Improvement Needed:**
+- Version bump should ALWAYS happen BEFORE tagging
+- **Make release.yml depend on validate-version.yml** using `needs:` or combine them
+- Consider using a reusable workflow or job dependency to enforce validation

README.md

@@ -24,114 +24,57 @@ Your AI coding companion. Build features, fix bugs, and ship faster — with aut
 - **Self-Validating**: Built-in QA loop catches issues before you review
 - **Isolated Workspaces**: All work happens in git worktrees — your code stays safe
 - **AI Merge Resolution**: Intelligent conflict resolution when merging back to main — no manual conflict fixing
-- **Memory Layer**: Agents remember insights across sessions for smarter decisions
 - **Cross-Platform**: Desktop app runs on Mac, Windows, and Linux
 - **Any Project Type**: Build web apps, APIs, CLIs — works with any software project
 
-## 🚀 Quick Start (Desktop UI)
+## Quick Start
 
-The Desktop UI is the recommended way to use Auto Claude. It provides visual task management, real-time progress tracking, and a Kanban board interface.
+### Download Auto Claude
+
+Download the latest release for your platform from [GitHub Releases](https://github.com/AndyMik90/Auto-Claude/releases/latest):
+
+| Platform | Download |
+|----------|----------|
+| **macOS (Apple Silicon M1-M4)** | `*-arm64.dmg` |
+| **macOS (Intel)** | `*-x64.dmg` |
+| **Windows** | `*.exe` |
+| **Linux** | `*.AppImage` or `*.deb` |
+
+> **Not sure which Mac?** Click the Apple menu () > "About This Mac". Look for "Chip" - M1/M2/M3/M4 = Apple Silicon, otherwise Intel.
 
 ### Prerequisites
 
-1. **Node.js 18+** - [Download Node.js](https://nodejs.org/)
-2. **Python 3.10+** - [Download Python](https://www.python.org/downloads/)
-3. **Docker Desktop** - Required for the Memory Layer
-4. **Claude Code CLI** - `npm install -g @anthropic-ai/claude-code`
-5. **Claude Subscription** - Requires [Claude Pro or Max](https://claude.ai/upgrade) for Claude Code access
-6. **Git Repository** - Your project must be initialized as a git repository
+Before using Auto Claude, you need:
 
-### Git Initialization
+1. **Claude Subscription** - Requires [Claude Pro or Max](https://claude.ai/upgrade) for Claude Code access
+2. **Claude Code CLI** - Install with: `npm install -g @anthropic-ai/claude-code`
 
-**Auto Claude requires a git repository** to create isolated worktrees for safe parallel development. If your project isn't a git repo yet:
+### Install and Run
 
-```bash
-cd your-project
-git init
-git add .
-git commit -m "Initial commit"
-```
-
-> **Why git?** Auto Claude uses git branches and worktrees to isolate each task in its own workspace, keeping your main branch clean until you're ready to merge. This allows you to work on multiple features simultaneously without conflicts.
-
----
-
-### Installing Docker Desktop
-
-Docker runs the FalkorDB database that powers Auto Claude's cross-session memory.
-
-| Operating System | Download Link |
-|------------------|---------------|
-| **Mac (Apple Silicon M1/M2/M3/M4)** | [Download for Apple Chip](https://desktop.docker.com/mac/main/arm64/Docker.dmg) |
-| **Mac (Intel)** | [Download for Intel Chip](https://desktop.docker.com/mac/main/amd64/Docker.dmg) |
-| **Windows** | [Download for Windows](https://desktop.docker.com/win/main/amd64/Docker%20Desktop%20Installer.exe) |
-| **Linux** | [Installation Guide](https://docs.docker.com/desktop/install/linux-install/) |
-
-> **Not sure which Mac?** Click the Apple menu (🍎) → "About This Mac". Look for "Chip" - M1/M2/M3/M4 = Apple Silicon, otherwise Intel.
-
-**After installing:** Open Docker Desktop and wait for the whale icon (🐳) to appear in your menu bar/system tray.
-
-> **Using the Desktop UI?** It automatically detects Docker status and offers one-click FalkorDB setup. No terminal commands needed!
-
-📚 **For detailed installation steps, troubleshooting, and advanced configuration, see [guides/DOCKER-SETUP.md](guides/DOCKER-SETUP.md)**
-
----
-
-### Step 1: Set Up the Python Backend
-
-The Desktop UI runs Python scripts behind the scenes. Set up the Python environment:
-
-```bash
-cd auto-claude
-
-# Using uv (recommended)
-uv venv && uv pip install -r requirements.txt
-
-# Or using standard Python
-python3 -m venv .venv && source .venv/bin/activate && pip install -r requirements.txt
-```
-
-### Step 2: Start the Memory Layer
-
-The Auto Claude Memory Layer provides cross-session context retention using a graph database:
-
-```bash
-# Make sure Docker Desktop is running, then:
-docker-compose up -d falkordb
-```
-
-### Step 3: Install and Launch the Desktop UI
-
-```bash
-cd auto-claude-ui
-
-# Install dependencies (pnpm recommended, npm works too)
-pnpm install
-# or: npm install
-
-# Build and start the application
-pnpm run build && pnpm run start
-# or: npm run build && npm run start
-```
+1. **Download** the installer for your platform from the table above
+2. **Install**:
+   - **macOS**: Open the `.dmg`, drag Auto Claude to Applications
+   - **Windows**: Run the `.exe` installer (see note below about security warning)
+   - **Linux**: Make the AppImage executable (`chmod +x`) and run it, or install the `.deb`
+3. **Launch** Auto Claude
+4. **Add your project** and start building!
 
 <details>
-<summary><b>Windows users:</b> If installation fails with node-gyp errors, click here</summary>
+<summary><b>Windows users:</b> Security warning when installing</summary>
 
-Auto Claude automatically downloads prebuilt binaries for Windows. If prebuilts aren't available for your Electron version yet, you'll need Visual Studio Build Tools:
+The Windows installer is not yet code-signed, so you may see a "Windows protected your PC" warning from Microsoft Defender SmartScreen.
 
-1. Download [Visual Studio Build Tools 2022](https://visualstudio.microsoft.com/visual-cpp-build-tools/)
-2. Select "Desktop development with C++" workload
-3. In "Individual Components", add "MSVC v143 - VS 2022 C++ x64/x86 Spectre-mitigated libs"
-4. Restart terminal and run `npm install` again
+**To proceed:**
+1. Click "More info"
+2. Click "Run anyway"
+
+This is safe — all releases are automatically scanned with VirusTotal before publishing. You can verify any installer by checking the **VirusTotal Scan Results** section in each [release's notes](https://github.com/AndyMik90/Auto-Claude/releases).
+
+We're working on obtaining a code signing certificate for future releases.
 
 </details>
 
-### Step 4: Start Building
-
-1. Add your project in the UI
-2. Create a new task describing what you want to build
-3. Watch as Auto Claude creates a spec, plans, and implements your feature
-4. Review changes and merge when satisfied
+> **Want to build from source?** See [CONTRIBUTING.md](CONTRIBUTING.md#running-from-source) for development setup.
 
 ---
 
@@ -241,22 +184,6 @@ Three-layer defense keeps your code safe:
 - **Filesystem Restrictions** — Operations limited to project directory
 - **Command Allowlist** — Only approved commands based on your project's stack
 
-### 🧠 Memory Layer
-
-The Memory Layer is a **hybrid RAG system** combining graph nodes with semantic search to deliver the best possible context during AI coding. Agents remember insights from previous sessions, discovered codebase patterns persist and are reusable, and historical context helps agents make smarter decisions.
-
-**Architecture:**
-- **Backend**: FalkorDB (graph database) via Docker
-- **Library**: Graphiti for knowledge graph operations
-- **Providers**: OpenAI, Anthropic, Azure OpenAI, or Ollama (local/offline)
-
-| Setup | LLM | Embeddings | Notes |
-|-------|-----|------------|-------|
-| **OpenAI** | OpenAI | OpenAI | Simplest - single API key |
-| **Anthropic + Voyage** | Anthropic | Voyage AI | High quality |
-| **Ollama** | Ollama | Ollama | Fully offline |
-| **Azure** | Azure OpenAI | Azure OpenAI | Enterprise |
-
 ## Project Structure
 
 ```
@@ -272,9 +199,8 @@ your-project/
 │   ├── spec_runner.py        # Spec creation orchestrator
 │   ├── prompts/              # Agent prompt templates
 │   └── ...
-├── auto-claude-ui/           # Electron desktop application
-│   └── ...
-└── docker-compose.yml        # FalkorDB for Memory Layer
+└── auto-claude-ui/           # Electron desktop application
+    └── ...
 ```
 
 ### Understanding the Folders
@@ -308,12 +234,6 @@ The `.auto-claude/` directory is gitignored and project-specific - you'll have o
 |----------|----------|-------------|
 | `CLAUDE_CODE_OAUTH_TOKEN` | Yes | OAuth token from `claude setup-token` |
 | `AUTO_BUILD_MODEL` | No | Model override (default: claude-opus-4-5-20251101) |
-| `GRAPHITI_ENABLED` | Recommended | Set to `true` to enable Memory Layer |
-| `GRAPHITI_LLM_PROVIDER` | For Memory | LLM provider: openai, anthropic, azure_openai, ollama |
-| `GRAPHITI_EMBEDDER_PROVIDER` | For Memory | Embedder: openai, voyage, azure_openai, ollama |
-| `OPENAI_API_KEY` | For OpenAI | Required for OpenAI provider |
-| `ANTHROPIC_API_KEY` | For Anthropic | Required for Anthropic LLM |
-| `VOYAGE_API_KEY` | For Voyage | Required for Voyage embeddings |
 
 See `auto-claude/.env.example` for complete configuration options.
 

RELEASE.md

@@ -0,0 +1,186 @@
+# Release Process
+
+This document describes how to create a new release of Auto Claude.
+
+## Automated Release Process (Recommended)
+
+We provide an automated script that handles version bumping, git commits, and tagging to ensure version consistency.
+
+### Prerequisites
+
+- Clean git working directory (no uncommitted changes)
+- You're on the branch you want to release from (usually `main`)
+
+### Steps
+
+1. **Run the version bump script:**
+
+   ```bash
+   # Bump patch version (2.5.5 -> 2.5.6)
+   node scripts/bump-version.js patch
+
+   # Bump minor version (2.5.5 -> 2.6.0)
+   node scripts/bump-version.js minor
+
+   # Bump major version (2.5.5 -> 3.0.0)
+   node scripts/bump-version.js major
+
+   # Set specific version
+   node scripts/bump-version.js 2.6.0
+   ```
+
+   This script will:
+   - ✅ Update `auto-claude-ui/package.json` with the new version
+   - ✅ Create a git commit with the version change
+   - ✅ Create a git tag (e.g., `v2.5.6`)
+   - ⚠️  **NOT** push to remote (you control when to push)
+
+2. **Review the changes:**
+
+   ```bash
+   git log -1              # View the commit
+   git show v2.5.6         # View the tag
+   ```
+
+3. **Push to GitHub:**
+
+   ```bash
+   # Push the commit
+   git push origin main
+
+   # Push the tag
+   git push origin v2.5.6
+   ```
+
+4. **Create GitHub Release:**
+
+   - Go to [GitHub Releases](https://github.com/AndyMik90/Auto-Claude/releases)
+   - Click "Draft a new release"
+   - Select the tag you just pushed (e.g., `v2.5.6`)
+   - Add release notes (describe what changed)
+   - Click "Publish release"
+
+5. **Automated builds will trigger:**
+
+   - ✅ Version validation workflow will verify version consistency
+   - ✅ Tests will run (`test-on-tag.yml`)
+   - ✅ Native module prebuilds will be created (`build-prebuilds.yml`)
+   - ✅ Discord notification will be sent (`discord-release.yml`)
+
+## Manual Release Process (Not Recommended)
+
+If you need to create a release manually, follow these steps **carefully** to avoid version mismatches:
+
+1. **Update `auto-claude-ui/package.json`:**
+
+   ```json
+   {
+     "version": "2.5.6"
+   }
+   ```
+
+2. **Commit the change:**
+
+   ```bash
+   git add auto-claude-ui/package.json
+   git commit -m "chore: bump version to 2.5.6"
+   ```
+
+3. **Create and push tag:**
+
+   ```bash
+   git tag -a v2.5.6 -m "Release v2.5.6"
+   git push origin main
+   git push origin v2.5.6
+   ```
+
+4. **Create GitHub Release** (same as step 4 above)
+
+## Version Validation
+
+A GitHub Action automatically validates that the version in `package.json` matches the git tag.
+
+If there's a mismatch, the workflow will **fail** with a clear error message:
+
+```
+❌ ERROR: Version mismatch detected!
+
+The version in package.json (2.5.0) does not match
+the git tag version (2.5.5).
+
+To fix this:
+  1. Delete this tag: git tag -d v2.5.5
+  2. Update package.json version to 2.5.5
+  3. Commit the change
+  4. Recreate the tag: git tag -a v2.5.5 -m 'Release v2.5.5'
+```
+
+This validation ensures we never ship a release where the updater shows the wrong version.
+
+## Troubleshooting
+
+### Version Mismatch Error
+
+If you see a version mismatch error in GitHub Actions:
+
+1. **Delete the incorrect tag:**
+   ```bash
+   git tag -d v2.5.6                    # Delete locally
+   git push origin :refs/tags/v2.5.6    # Delete remotely
+   ```
+
+2. **Use the automated script:**
+   ```bash
+   node scripts/bump-version.js 2.5.6
+   git push origin main
+   git push origin v2.5.6
+   ```
+
+### Git Working Directory Not Clean
+
+If the version bump script fails with "Git working directory is not clean":
+
+```bash
+# Commit or stash your changes first
+git status
+git add .
+git commit -m "your changes"
+
+# Then run the version bump script
+node scripts/bump-version.js patch
+```
+
+## Release Checklist
+
+Use this checklist when creating a new release:
+
+- [ ] All tests passing on main branch
+- [ ] CHANGELOG updated (if applicable)
+- [ ] Run `node scripts/bump-version.js <type>`
+- [ ] Review commit and tag
+- [ ] Push commit and tag to GitHub
+- [ ] Create GitHub Release with release notes
+- [ ] Verify version validation passed
+- [ ] Verify builds completed successfully
+- [ ] Test the updater shows correct version
+
+## What Gets Released
+
+When you create a release, the following are built and published:
+
+1. **Native module prebuilds** - Windows node-pty binaries
+2. **Electron app packages** - Desktop installers (triggered manually or via electron-builder)
+3. **Discord notification** - Sent to the Auto Claude community
+
+## Version Numbering
+
+We follow [Semantic Versioning (SemVer)](https://semver.org/):
+
+- **MAJOR** version (X.0.0) - Breaking changes
+- **MINOR** version (0.X.0) - New features (backward compatible)
+- **PATCH** version (0.0.X) - Bug fixes (backward compatible)
+
+Examples:
+- `2.5.5 -> 2.5.6` - Bug fix
+- `2.5.6 -> 2.6.0` - New feature
+- `2.6.0 -> 3.0.0` - Breaking change

agpl-3.0.txt

@@ -658,4 +658,4 @@ specific requirements.
   You should also get your employer (if you work as a programmer) or school,
 if any, to sign a "copyright disclaimer" for the program, if necessary.
 For more information on this, and how to apply and follow the GNU AGPL, see
-<https://www.gnu.org/licenses/>.
+<https://www.gnu.org/licenses/>.

auto-claude-ui/.env.example

@@ -5,23 +5,20 @@
 # DEBUG SETTINGS
 # ============================================
 
-# Enable general debug logging for ideation and roadmap features
+# Enable debug logging across the entire application
 # When enabled, you'll see detailed console logs for:
-# - Ideation generation and stop functionality
-# - Roadmap generation and stop functionality
+# - Ideation and roadmap generation
 # - IPC communication between processes
 # - Store state updates
+# - Changelog generation and project initialization
+# - GitHub OAuth flow
 # Usage: Set to 'true' before starting the app
 # DEBUG=true
 
-# Enable debug logging for the auto-updater
+# Enable debug logging for the auto-updater only
 # Shows detailed information about app update checks and downloads
 # DEBUG_UPDATER=true
 
-# Enable debug logging for Auto Claude features
-# Affects changelog generation, project initialization, and other core features
-# AUTO_CLAUDE_DEBUG=true
-
 # ============================================
 # HOW TO USE
 # ============================================
@@ -31,7 +28,6 @@
 #
 # Option 2: Export in your shell profile (~/.bashrc, ~/.zshrc, etc.)
 #   export DEBUG=true
-#   export AUTO_CLAUDE_DEBUG=true
 #
 # Option 3: Create a .env file in this directory (auto-claude-ui/)
 #   Copy this file: cp .env.example .env

auto-claude-ui/design.json

@@ -63,10 +63,10 @@
         "id": "default",
         "name": "Default",
         "description": "Oscura Midnight - deepest dark with saturated yellow accent, inspired by Fey/Oscura",
-        "previewColors": { 
-          "lightBg": "#F2F2ED", 
+        "previewColors": {
+          "lightBg": "#F2F2ED",
           "lightAccent": "#A5A66A",
-          "darkBg": "#0B0B0F", 
+          "darkBg": "#0B0B0F",
           "darkAccent": "#D6D876"
         },
         "semanticColors": {
@@ -81,10 +81,10 @@
         "id": "dusk",
         "name": "Dusk",
         "description": "Warmer Oscura variant with slightly lighter dark mode",
-        "previewColors": { 
-          "lightBg": "#F5F5F0", 
+        "previewColors": {
+          "lightBg": "#F5F5F0",
           "lightAccent": "#B8B978",
-          "darkBg": "#131419", 
+          "darkBg": "#131419",
           "darkAccent": "#E6E7A3"
         },
         "semanticColors": {
@@ -99,50 +99,50 @@
         "id": "lime",
         "name": "Lime",
         "description": "Fresh, energetic lime/chartreuse with purple accents",
-        "previewColors": { 
-          "lightBg": "#E8F5A3", 
-          "darkBg": "#0F0F1A", 
-          "accent": "#7C3AED" 
+        "previewColors": {
+          "lightBg": "#E8F5A3",
+          "darkBg": "#0F0F1A",
+          "accent": "#7C3AED"
         }
       },
       {
         "id": "ocean",
-        "name": "Ocean", 
+        "name": "Ocean",
         "description": "Calm, professional blue tones",
-        "previewColors": { 
-          "lightBg": "#E0F2FE", 
-          "darkBg": "#082F49", 
-          "accent": "#0284C7" 
+        "previewColors": {
+          "lightBg": "#E0F2FE",
+          "darkBg": "#082F49",
+          "accent": "#0284C7"
         }
       },
       {
         "id": "retro",
         "name": "Retro",
         "description": "Warm, nostalgic amber/orange vibes",
-        "previewColors": { 
-          "lightBg": "#FEF3C7", 
-          "darkBg": "#1C1917", 
-          "accent": "#D97706" 
+        "previewColors": {
+          "lightBg": "#FEF3C7",
+          "darkBg": "#1C1917",
+          "accent": "#D97706"
         }
       },
       {
         "id": "neo",
         "name": "Neo",
         "description": "Modern cyberpunk pink/magenta",
-        "previewColors": { 
-          "lightBg": "#FDF4FF", 
-          "darkBg": "#0F0720", 
-          "accent": "#D946EF" 
+        "previewColors": {
+          "lightBg": "#FDF4FF",
+          "darkBg": "#0F0720",
+          "accent": "#D946EF"
         }
       },
       {
         "id": "forest",
         "name": "Forest",
         "description": "Natural, earthy green tones",
-        "previewColors": { 
-          "lightBg": "#DCFCE7", 
-          "darkBg": "#052E16", 
-          "accent": "#16A34A" 
+        "previewColors": {
+          "lightBg": "#DCFCE7",
+          "darkBg": "#052E16",
+          "accent": "#16A34A"
         }
       }
     ],
@@ -152,7 +152,7 @@
   "colors": {
     "note": "These are the Default theme colors (Oscura Midnight). See themeSystem for all available themes.",
     "cssVariablePrefix": "--color-",
-    
+
     "lightMode": {
       "background": {
         "primary": "#F2F2ED",
@@ -186,7 +186,7 @@
         "focus": "#A5A66A"
       }
     },
-    
+
     "darkMode": {
       "background": {
         "primary": "#0B0B0F",
@@ -223,7 +223,7 @@
         "focus": "#D6D876"
       }
     },
-    
+
     "semantic": {
       "success": "#4EBE96",
       "successLight": { "light": "#E0F5ED", "dark": "#1A2924" },
@@ -238,7 +238,7 @@
       "infoLight": { "light": "#E8F4FF", "dark": "#1A2230" },
       "infoDescription": "Blue - for links and informational elements"
     },
-    
+
     "shadows": {
       "lightMode": {
         "sm": "0 1px 2px 0 rgba(0, 0, 0, 0.05)",
@@ -511,30 +511,30 @@
         "fontWeight": "500"
       },
       "variants": {
-        "default": { 
-          "background": "var(--color-background-secondary)", 
-          "text": "var(--color-text-secondary)" 
+        "default": {
+          "background": "var(--color-background-secondary)",
+          "text": "var(--color-text-secondary)"
         },
-        "primary": { 
-          "background": "var(--color-accent-primary-light)", 
-          "text": "var(--color-accent-primary)" 
+        "primary": {
+          "background": "var(--color-accent-primary-light)",
+          "text": "var(--color-accent-primary)"
         },
-        "success": { 
-          "background": "var(--color-semantic-success-light)", 
-          "text": "var(--color-semantic-success)" 
+        "success": {
+          "background": "var(--color-semantic-success-light)",
+          "text": "var(--color-semantic-success)"
         },
-        "warning": { 
-          "background": "var(--color-semantic-warning-light)", 
-          "text": "var(--color-semantic-warning)" 
+        "warning": {
+          "background": "var(--color-semantic-warning-light)",
+          "text": "var(--color-semantic-warning)"
         },
-        "error": { 
-          "background": "var(--color-semantic-error-light)", 
-          "text": "var(--color-semantic-error)" 
+        "error": {
+          "background": "var(--color-semantic-error-light)",
+          "text": "var(--color-semantic-error)"
         },
-        "outline": { 
-          "background": "transparent", 
-          "border": "1px solid var(--color-border-default)", 
-          "text": "var(--color-text-secondary)" 
+        "outline": {
+          "background": "transparent",
+          "border": "1px solid var(--color-border-default)",
+          "text": "var(--color-text-secondary)"
         }
       }
     },
@@ -616,10 +616,10 @@
       "description": "Date picker grid with clear day cells and selection states.",
       "styling": {
         "dayCell": { "size": "36px", "borderRadius": "md (8px)" },
-        "selectedDay": { 
-          "background": "var(--color-accent-primary)", 
-          "text": "var(--color-text-inverse)", 
-          "borderRadius": "full" 
+        "selectedDay": {
+          "background": "var(--color-accent-primary)",
+          "text": "var(--color-text-inverse)",
+          "borderRadius": "full"
         },
         "todayIndicator": "var(--color-accent-primary) text color or dot",
         "rangeSelection": "var(--color-accent-primary-light) background for range days"
@@ -634,14 +634,14 @@
         "thumbSize": "20px"
       },
       "styling": {
-        "off": { 
-          "track": "var(--color-border-default)", 
+        "off": {
+          "track": "var(--color-border-default)",
           "lightTrack": "#DEDED9",
           "darkTrack": "#232323",
-          "thumb": "white" 
+          "thumb": "white"
         },
-        "on": { 
-          "track": "var(--color-accent-primary)", 
+        "on": {
+          "track": "var(--color-accent-primary)",
           "lightTrack": "#A5A66A",
           "darkTrack": "#D6D876",
           "thumb": "var(--color-text-inverse)",
@@ -895,7 +895,7 @@
     "cssVariableMap": {
       "backgrounds": [
         "--color-background-primary",
-        "--color-background-secondary", 
+        "--color-background-secondary",
         "--color-background-neutral"
       ],
       "surfaces": [

auto-claude-ui/electron.vite.config.ts

@@ -9,7 +9,7 @@ export default defineConfig({
       exclude: [
         'uuid',
         'chokidar',
-        'ioredis',
+        'kuzu',
         'electron-updater',
         '@electron-toolkit/utils'
       ]
@@ -20,7 +20,7 @@ export default defineConfig({
           index: resolve(__dirname, 'src/main/index.ts')
         },
         // Only node-pty needs to be external (native module rebuilt by electron-builder)
-        external: ['node-pty']
+        external: ['@lydell/node-pty']
       }
     }
   },

auto-claude-ui/package.json

@@ -1,22 +1,32 @@
 {
   "name": "auto-claude-ui",
-  "version": "2.5.0",
+  "version": "2.7.1",
   "description": "Desktop UI for Auto Claude autonomous coding framework",
+  "homepage": "https://github.com/AndyMik90/Auto-Claude",
+  "repository": {
+    "type": "git",
+    "url": "https://github.com/AndyMik90/Auto-Claude.git"
+  },
   "main": "./out/main/index.js",
-  "author": "Auto Claude Team",
+  "author": {
+    "name": "Auto Claude Team",
+    "email": "119136210+AndyMik90@users.noreply.github.com"
+  },
   "license": "AGPL-3.0",
   "scripts": {
     "postinstall": "node scripts/postinstall.js",
     "dev": "electron-vite dev",
+    "dev:mcp": "electron-vite dev -- --remote-debugging-port=9222",
     "build": "electron-vite build",
     "start": "electron .",
+    "start:mcp": "electron . --remote-debugging-port=9222",
     "preview": "electron-vite preview",
     "package": "electron-vite build && electron-builder",
     "package:mac": "electron-vite build && electron-builder --mac",
     "package:win": "electron-vite build && electron-builder --win",
     "package:linux": "electron-vite build && electron-builder --linux",
-    "start:packaged:mac": "open dist/mac-arm64/Auto\\ Claude.app || open dist/mac/Auto\\ Claude.app",
-    "start:packaged:win": "start \"\" \"dist\\win-unpacked\\Auto Claude.exe\"",
+    "start:packaged:mac": "open dist/mac-arm64/Auto-Claude.app || open dist/mac/Auto-Claude.app",
+    "start:packaged:win": "start \"\" \"dist\\win-unpacked\\Auto-Claude.exe\"",
     "start:packaged:linux": "./dist/linux-unpacked/auto-claude",
     "test": "vitest run",
     "test:watch": "vitest",
@@ -30,6 +40,7 @@
     "@dnd-kit/core": "^6.3.1",
     "@dnd-kit/sortable": "^10.0.0",
     "@dnd-kit/utilities": "^3.2.2",
+    "@lydell/node-pty": "^1.1.0",
     "@radix-ui/react-alert-dialog": "^1.1.15",
     "@radix-ui/react-checkbox": "^1.1.4",
     "@radix-ui/react-collapsible": "^1.1.3",
@@ -56,10 +67,8 @@
     "class-variance-authority": "^0.7.1",
     "clsx": "^2.1.1",
     "electron-updater": "^6.6.2",
-    "ioredis": "^5.8.2",
     "lucide-react": "^0.560.0",
     "motion": "^12.23.26",
-    "node-pty": "^1.1.0-beta42",
     "react": "^19.2.3",
     "react-dom": "^19.2.3",
     "react-markdown": "^10.1.0",
@@ -77,7 +86,6 @@
     "@playwright/test": "^1.52.0",
     "@tailwindcss/postcss": "^4.1.17",
     "@testing-library/react": "^16.1.0",
-    "@types/ioredis": "^4.28.10",
     "@types/node": "^25.0.0",
     "@types/react": "^19.2.7",
     "@types/react-dom": "^19.2.3",
@@ -104,17 +112,20 @@
   "pnpm": {
     "overrides": {
       "electron-builder-squirrel-windows": "^26.0.12",
-      "dmg-builder": "^26.0.12"
+      "dmg-builder": "^26.0.12",
+      "node-pty": "npm:@lydell/node-pty@^1.1.0"
     },
     "onlyBuiltDependencies": [
+      "@lydell/node-pty",
       "electron",
-      "esbuild",
-      "node-pty"
+      "electron-winstaller",
+      "esbuild"
     ]
   },
   "build": {
     "appId": "com.autoclaude.ui",
-    "productName": "Auto Claude",
+    "productName": "Auto-Claude",
+    "artifactName": "${productName}-${version}-${platform}-${arch}.${ext}",
     "publish": [
       {
         "provider": "github",
@@ -132,8 +143,8 @@
     ],
     "extraResources": [
       {
-        "from": "node_modules/node-pty",
-        "to": "node_modules/node-pty"
+        "from": "node_modules/@lydell/node-pty",
+        "to": "node_modules/@lydell/node-pty"
       },
       {
         "from": "resources/icon.ico",
@@ -148,13 +159,23 @@
           "!**/*.pyc",
           "!**/specs",
           "!**/.venv",
-          "!**/.env"
+          "!**/.venv-*",
+          "!**/venv",
+          "!**/.env",
+          "!**/tests",
+          "!**/*.egg-info",
+          "!**/.pytest_cache",
+          "!**/.mypy_cache"
         ]
       }
     ],
     "mac": {
       "category": "public.app-category.developer-tools",
       "icon": "resources/icon.icns",
+      "hardenedRuntime": true,
+      "gatekeeperAssess": false,
+      "entitlements": "resources/entitlements.mac.plist",
+      "entitlementsInherit": "resources/entitlements.mac.plist",
       "target": [
         "dmg",
         "zip"
@@ -180,5 +201,6 @@
     "*.{ts,tsx}": [
       "eslint --fix"
     ]
-  }
+  },
+  "packageManager": "pnpm@10.26.1+sha512.664074abc367d2c9324fdc18037097ce0a8f126034160f709928e9e9f95d98714347044e5c3164d65bd5da6c59c6be362b107546292a8eecb7999196e5ce58fa"
 }

auto-claude-ui/resources/entitlements.mac.plist

@@ -0,0 +1,24 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
+<plist version="1.0">
+<dict>
+    <!-- Allow the app to be debugged -->
+    <key>com.apple.security.cs.allow-jit</key>
+    <true/>
+    <!-- Allow loading unsigned libraries (needed for native modules) -->
+    <key>com.apple.security.cs.allow-unsigned-executable-memory</key>
+    <true/>
+    <!-- Allow dyld environment variables (needed for Electron) -->
+    <key>com.apple.security.cs.disable-library-validation</key>
+    <true/>
+    <!-- Allow spawning child processes (needed for Python backend and terminals) -->
+    <key>com.apple.security.cs.allow-dyld-environment-variables</key>
+    <true/>
+    <!-- Network access for API calls -->
+    <key>com.apple.security.network.client</key>
+    <true/>
+    <!-- File access for user documents -->
+    <key>com.apple.security.files.user-selected.read-write</key>
+    <true/>
+</dict>
+</plist>

auto-claude-ui/src/__tests__/integration/subprocess-spawn.test.ts

@@ -6,11 +6,16 @@ import { describe, it, expect, vi, beforeEach, afterEach } from 'vitest';
 import { EventEmitter } from 'events';
 import { mkdirSync, rmSync, existsSync, writeFileSync } from 'fs';
 import path from 'path';
+import { findPythonCommand, parsePythonCommand } from '../../main/python-detector';
 
 // Test directories
 const TEST_DIR = '/tmp/subprocess-spawn-test';
 const TEST_PROJECT_PATH = path.join(TEST_DIR, 'test-project');
 
+// Detect the Python command that will actually be used
+const DETECTED_PYTHON_CMD = findPythonCommand() || 'python';
+const [EXPECTED_PYTHON_COMMAND, EXPECTED_PYTHON_BASE_ARGS] = parsePythonCommand(DETECTED_PYTHON_CMD);
+
 // Mock child_process spawn
 const mockStdout = new EventEmitter();
 const mockStderr = new EventEmitter();
@@ -99,8 +104,9 @@ describe('Subprocess Spawn Integration', () => {
       manager.startSpecCreation('task-1', TEST_PROJECT_PATH, 'Test task description');
 
       expect(spawn).toHaveBeenCalledWith(
-        'python3',
+        EXPECTED_PYTHON_COMMAND,
         expect.arrayContaining([
+          ...EXPECTED_PYTHON_BASE_ARGS,
           expect.stringContaining('spec_runner.py'),
           '--task',
           'Test task description'
@@ -123,8 +129,13 @@ describe('Subprocess Spawn Integration', () => {
       manager.startTaskExecution('task-1', TEST_PROJECT_PATH, 'spec-001');
 
       expect(spawn).toHaveBeenCalledWith(
-        'python3',
-        expect.arrayContaining([expect.stringContaining('run.py'), '--spec', 'spec-001']),
+        EXPECTED_PYTHON_COMMAND,
+        expect.arrayContaining([
+          ...EXPECTED_PYTHON_BASE_ARGS,
+          expect.stringContaining('run.py'),
+          '--spec',
+          'spec-001'
+        ]),
         expect.objectContaining({
           cwd: AUTO_CLAUDE_SOURCE  // Process runs from auto-claude source directory
         })
@@ -140,8 +151,9 @@ describe('Subprocess Spawn Integration', () => {
       manager.startQAProcess('task-1', TEST_PROJECT_PATH, 'spec-001');
 
       expect(spawn).toHaveBeenCalledWith(
-        'python3',
+        EXPECTED_PYTHON_COMMAND,
         expect.arrayContaining([
+          ...EXPECTED_PYTHON_BASE_ARGS,
           expect.stringContaining('run.py'),
           '--spec',
           'spec-001',
@@ -167,8 +179,13 @@ describe('Subprocess Spawn Integration', () => {
 
       // Should spawn normally - parallel options don't affect CLI args anymore
       expect(spawn).toHaveBeenCalledWith(
-        'python3',
-        expect.arrayContaining([expect.stringContaining('run.py'), '--spec', 'spec-001']),
+        EXPECTED_PYTHON_COMMAND,
+        expect.arrayContaining([
+          ...EXPECTED_PYTHON_BASE_ARGS,
+          expect.stringContaining('run.py'),
+          '--spec',
+          'spec-001'
+        ]),
         expect.any(Object)
       );
     });

auto-claude-ui/src/__tests__/setup.ts

@@ -5,11 +5,37 @@ import { vi, beforeEach, afterEach } from 'vitest';
 import { mkdirSync, rmSync, existsSync } from 'fs';
 import path from 'path';
 
+// Mock localStorage for tests that need it
+const localStorageMock = (() => {
+  let store: Record<string, string> = {};
+
+  return {
+    getItem: vi.fn((key: string) => store[key] || null),
+    setItem: vi.fn((key: string, value: string) => {
+      store[key] = value;
+    }),
+    removeItem: vi.fn((key: string) => {
+      delete store[key];
+    }),
+    clear: vi.fn(() => {
+      store = {};
+    })
+  };
+})();
+
+// Make localStorage available globally
+Object.defineProperty(global, 'localStorage', {
+  value: localStorageMock
+});
+
 // Test data directory for isolated file operations
 export const TEST_DATA_DIR = '/tmp/auto-claude-ui-tests';
 
 // Create fresh test directory before each test
 beforeEach(() => {
+  // Clear localStorage
+  localStorageMock.clear();
+
   // Use a unique subdirectory per test to avoid race conditions in parallel tests
   const testId = `test-${Date.now()}-${Math.random().toString(36).slice(2, 8)}`;
   const _testDir = path.join(TEST_DATA_DIR, testId);
@@ -56,7 +82,13 @@ if (typeof window !== 'undefined') {
     getSettings: vi.fn(),
     saveSettings: vi.fn(),
     selectDirectory: vi.fn(),
-    getAppVersion: vi.fn()
+    getAppVersion: vi.fn(),
+    // Tab state persistence (IPC-based)
+    getTabState: vi.fn().mockResolvedValue({
+      success: true,
+      data: { openProjectIds: [], activeProjectId: null, tabOrder: [] }
+    }),
+    saveTabState: vi.fn().mockResolvedValue({ success: true })
   };
 }
 

auto-claude-ui/src/main/__tests__/ipc-handlers.test.ts

@@ -39,6 +39,14 @@ vi.mock('@electron-toolkit/utils', () => ({
   }
 }));
 
+// Mock version-manager to return a predictable version
+vi.mock('../updater/version-manager', () => ({
+  getEffectiveVersion: vi.fn(() => '0.1.0'),
+  getBundledVersion: vi.fn(() => '0.1.0'),
+  parseVersionFromTag: vi.fn((tag: string) => tag.replace('v', '')),
+  compareVersions: vi.fn(() => 0)
+}));
+
 // Mock modules before importing
 vi.mock('electron', () => {
   const mockIpcMain = new (class extends EventEmitter {

auto-claude-ui/src/main/__tests__/rate-limit-auto-recovery.test.ts

@@ -0,0 +1,605 @@
+/**
+ * Integration tests for Rate Limit Auto-Recovery System
+ * Tests the complete flow: rate limit detection → account swap → task restart
+ */
+import { describe, it, expect, vi, beforeEach, afterEach } from 'vitest';
+import { EventEmitter } from 'events';
+
+// Mock data
+const mockProfiles = {
+  mai: {
+    id: 'profile-mai',
+    name: 'MAI',
+    email: 'mai@example.com',
+    isDefault: true,
+    oauthToken: 'encrypted-token-mai',
+    createdAt: new Date(),
+    rateLimitEvents: []
+  },
+  mu: {
+    id: 'profile-mu',
+    name: 'MU',
+    email: 'mu@example.com',
+    isDefault: false,
+    oauthToken: 'encrypted-token-mu',
+    createdAt: new Date(),
+    rateLimitEvents: []
+  }
+};
+
+const mockAutoSwitchSettings = {
+  enabled: true,
+  proactiveSwapEnabled: true,
+  sessionThreshold: 95,
+  weeklyThreshold: 99,
+  autoSwitchOnRateLimit: true,
+  usageCheckInterval: 30000
+};
+
+// Create mock profile manager
+function createMockProfileManager(options: {
+  activeProfileId?: string;
+  profiles?: typeof mockProfiles;
+  autoSwitchSettings?: typeof mockAutoSwitchSettings;
+  bestAvailableProfile?: typeof mockProfiles.mai | null;
+} = {}) {
+  const activeId = options.activeProfileId || 'profile-mai';
+  const profiles = options.profiles || mockProfiles;
+  const settings = options.autoSwitchSettings || mockAutoSwitchSettings;
+  const bestProfile = options.bestAvailableProfile !== undefined
+    ? options.bestAvailableProfile
+    : profiles.mu;
+
+  return {
+    getActiveProfile: vi.fn(() => profiles[activeId === 'profile-mai' ? 'mai' : 'mu']),
+    getProfile: vi.fn((id: string) => {
+      if (id === 'profile-mai') return profiles.mai;
+      if (id === 'profile-mu') return profiles.mu;
+      return null;
+    }),
+    getBestAvailableProfile: vi.fn((_excludeProfileId?: string) => bestProfile),
+    setActiveProfile: vi.fn(),
+    recordRateLimitEvent: vi.fn(),
+    getAutoSwitchSettings: vi.fn(() => settings),
+    getProfileToken: vi.fn(() => 'decrypted-token'),
+    getActiveProfileToken: vi.fn(() => 'decrypted-token')
+  };
+}
+
+describe('Rate Limit Auto-Recovery Integration', () => {
+  let mockProfileManager: ReturnType<typeof createMockProfileManager>;
+
+  beforeEach(() => {
+    vi.resetModules();
+    mockProfileManager = createMockProfileManager();
+  });
+
+  afterEach(() => {
+    vi.clearAllMocks();
+  });
+
+  describe('Rate Limit Detection Patterns', () => {
+    beforeEach(() => {
+      vi.doMock('../claude-profile-manager', () => ({
+        getClaudeProfileManager: vi.fn(() => mockProfileManager)
+      }));
+    });
+
+    it('should detect standard Claude rate limit message', async () => {
+      const { detectRateLimit } = await import('../rate-limit-detector');
+
+      const output = 'Limit reached · resets Dec 17 at 6am (Europe/Oslo)';
+      const result = detectRateLimit(output);
+
+      expect(result.isRateLimited).toBe(true);
+      expect(result.resetTime).toBe('Dec 17 at 6am (Europe/Oslo)');
+      expect(result.limitType).toBe('weekly');
+    });
+
+    it('should detect session limit (time only reset)', async () => {
+      const { detectRateLimit } = await import('../rate-limit-detector');
+
+      const output = 'Limit reached • resets 11:59pm';
+      const result = detectRateLimit(output);
+
+      expect(result.isRateLimited).toBe(true);
+      expect(result.limitType).toBe('session');
+    });
+
+    it('should detect rate limit in multiline output', async () => {
+      const { detectRateLimit } = await import('../rate-limit-detector');
+
+      const output = `Processing task...
+Some output here
+Limit reached · resets Dec 20 at 3pm (America/New_York)
+Stack trace follows`;
+
+      const result = detectRateLimit(output);
+
+      expect(result.isRateLimited).toBe(true);
+      expect(result.resetTime).toBe('Dec 20 at 3pm (America/New_York)');
+    });
+
+    it('should suggest alternative profile when rate limited', async () => {
+      const { detectRateLimit } = await import('../rate-limit-detector');
+
+      const output = 'Limit reached · resets Dec 17 at 6am';
+      const result = detectRateLimit(output, 'profile-mai');
+
+      expect(result.isRateLimited).toBe(true);
+      expect(result.suggestedProfile).toBeDefined();
+      expect(result.suggestedProfile?.id).toBe('profile-mu');
+    });
+
+    it('should record rate limit event in profile manager', async () => {
+      const { detectRateLimit } = await import('../rate-limit-detector');
+
+      detectRateLimit('Limit reached · resets Dec 17 at 6am', 'profile-mai');
+
+      expect(mockProfileManager.recordRateLimitEvent).toHaveBeenCalledWith(
+        'profile-mai',
+        'Dec 17 at 6am'
+      );
+    });
+  });
+
+  describe('Auto-Switch Settings Verification', () => {
+    it('should respect enabled flag', async () => {
+      const disabledManager = createMockProfileManager({
+        autoSwitchSettings: { ...mockAutoSwitchSettings, enabled: false }
+      });
+
+      vi.doMock('../claude-profile-manager', () => ({
+        getClaudeProfileManager: vi.fn(() => disabledManager)
+      }));
+
+      const settings = disabledManager.getAutoSwitchSettings();
+
+      expect(settings.enabled).toBe(false);
+      // When enabled is false, auto-swap should NOT happen even if autoSwitchOnRateLimit is true
+    });
+
+    it('should respect autoSwitchOnRateLimit flag', async () => {
+      const manualManager = createMockProfileManager({
+        autoSwitchSettings: { ...mockAutoSwitchSettings, autoSwitchOnRateLimit: false }
+      });
+
+      const settings = manualManager.getAutoSwitchSettings();
+
+      expect(settings.enabled).toBe(true);
+      expect(settings.autoSwitchOnRateLimit).toBe(false);
+      // When autoSwitchOnRateLimit is false, should show manual modal instead
+    });
+
+    it('should have both enabled and autoSwitchOnRateLimit for auto-recovery', () => {
+      const settings = mockProfileManager.getAutoSwitchSettings();
+
+      // Both must be true for automatic recovery
+      const shouldAutoRecover = settings.enabled && settings.autoSwitchOnRateLimit;
+      expect(shouldAutoRecover).toBe(true);
+    });
+  });
+
+  describe('Profile Scoring and Selection', () => {
+    it('should return alternative profile when one is available', () => {
+      const bestProfile = mockProfileManager.getBestAvailableProfile('profile-mai');
+
+      expect(bestProfile).toBeDefined();
+      expect(bestProfile?.id).toBe('profile-mu');
+    });
+
+    it('should return null when no alternative profile is available', () => {
+      const noAlternativeManager = createMockProfileManager({
+        bestAvailableProfile: null
+      });
+
+      const bestProfile = noAlternativeManager.getBestAvailableProfile('profile-mai');
+
+      expect(bestProfile).toBeNull();
+    });
+
+    it('should not return the same profile that hit the limit', () => {
+      const bestProfile = mockProfileManager.getBestAvailableProfile('profile-mai');
+
+      // Best profile should be different from the one that hit the limit
+      expect(bestProfile?.id).not.toBe('profile-mai');
+    });
+  });
+
+  describe('Auto-Recovery Flow Simulation', () => {
+    /**
+     * Simulates the flow in agent-process.ts lines 274-327
+     */
+    function simulateRateLimitRecovery(
+      output: string,
+      exitCode: number,
+      profileManager: ReturnType<typeof createMockProfileManager>
+    ): {
+      rateLimitDetected: boolean;
+      autoSwapped: boolean;
+      taskRestarted: boolean;
+      modalShown: boolean;
+      swappedToProfile?: { id: string; name: string };
+    } {
+      const result = {
+        rateLimitDetected: false,
+        autoSwapped: false,
+        taskRestarted: false,
+        modalShown: false,
+        swappedToProfile: undefined as { id: string; name: string } | undefined
+      };
+
+      // Only check rate limit if process failed
+      if (exitCode !== 0) {
+        // Simulate detectRateLimit
+        const rateLimitPattern = /Limit reached\s*[·•]\s*resets\s+(.+?)(?:\s*$|\n)/im;
+        const rateIndicators = [/rate\s*limit/i, /usage\s*limit/i, /limit\s*reached/i];
+
+        const isRateLimited = rateLimitPattern.test(output) ||
+          rateIndicators.some(p => p.test(output));
+
+        if (isRateLimited) {
+          result.rateLimitDetected = true;
+
+          const settings = profileManager.getAutoSwitchSettings();
+
+          if (settings.enabled && settings.autoSwitchOnRateLimit) {
+            const bestProfile = profileManager.getBestAvailableProfile('current-profile');
+
+            if (bestProfile) {
+              // Auto-swap
+              profileManager.setActiveProfile(bestProfile.id);
+              result.autoSwapped = true;
+              result.swappedToProfile = { id: bestProfile.id, name: bestProfile.name };
+              result.taskRestarted = true;
+              result.modalShown = true; // Notification modal
+            } else {
+              // No alternative - show manual modal
+              result.modalShown = true;
+            }
+          } else {
+            // Auto-switch disabled - show manual modal
+            result.modalShown = true;
+          }
+        }
+      }
+
+      return result;
+    }
+
+    it('should auto-swap and restart when all conditions met', () => {
+      const result = simulateRateLimitRecovery(
+        'Limit reached · resets Dec 17 at 6am',
+        1, // non-zero exit
+        mockProfileManager
+      );
+
+      expect(result.rateLimitDetected).toBe(true);
+      expect(result.autoSwapped).toBe(true);
+      expect(result.taskRestarted).toBe(true);
+      expect(result.modalShown).toBe(true);
+      expect(result.swappedToProfile?.id).toBe('profile-mu');
+    });
+
+    it('should NOT auto-swap when exit code is 0', () => {
+      const result = simulateRateLimitRecovery(
+        'Limit reached · resets Dec 17 at 6am',
+        0, // success exit
+        mockProfileManager
+      );
+
+      expect(result.rateLimitDetected).toBe(false);
+      expect(result.autoSwapped).toBe(false);
+      expect(result.taskRestarted).toBe(false);
+    });
+
+    it('should NOT auto-swap when enabled is false', () => {
+      const disabledManager = createMockProfileManager({
+        autoSwitchSettings: { ...mockAutoSwitchSettings, enabled: false }
+      });
+
+      const result = simulateRateLimitRecovery(
+        'Limit reached · resets Dec 17 at 6am',
+        1,
+        disabledManager
+      );
+
+      expect(result.rateLimitDetected).toBe(true);
+      expect(result.autoSwapped).toBe(false);
+      expect(result.modalShown).toBe(true); // Manual modal
+    });
+
+    it('should NOT auto-swap when autoSwitchOnRateLimit is false', () => {
+      const manualManager = createMockProfileManager({
+        autoSwitchSettings: { ...mockAutoSwitchSettings, autoSwitchOnRateLimit: false }
+      });
+
+      const result = simulateRateLimitRecovery(
+        'Limit reached · resets Dec 17 at 6am',
+        1,
+        manualManager
+      );
+
+      expect(result.rateLimitDetected).toBe(true);
+      expect(result.autoSwapped).toBe(false);
+      expect(result.modalShown).toBe(true); // Manual modal
+    });
+
+    it('should show manual modal when no alternative profile available', () => {
+      const noAlternativeManager = createMockProfileManager({
+        bestAvailableProfile: null
+      });
+
+      const result = simulateRateLimitRecovery(
+        'Limit reached · resets Dec 17 at 6am',
+        1,
+        noAlternativeManager
+      );
+
+      expect(result.rateLimitDetected).toBe(true);
+      expect(result.autoSwapped).toBe(false);
+      expect(result.taskRestarted).toBe(false);
+      expect(result.modalShown).toBe(true); // Manual modal because no alternative
+    });
+
+    it('should NOT detect rate limit for normal errors', () => {
+      const result = simulateRateLimitRecovery(
+        'Error: File not found',
+        1,
+        mockProfileManager
+      );
+
+      expect(result.rateLimitDetected).toBe(false);
+      expect(result.autoSwapped).toBe(false);
+      expect(result.modalShown).toBe(false);
+    });
+  });
+
+  describe('Task Restart Context Preservation', () => {
+    it('should preserve task context for restart', () => {
+      // Simulate task execution context
+      const taskContext = {
+        taskId: 'task-123',
+        projectPath: '/path/to/project',
+        specId: 'spec-001',
+        options: { qa: false },
+        swapCount: 0,
+        isSpecCreation: false
+      };
+
+      // After swap, swapCount should increment
+      taskContext.swapCount++;
+
+      expect(taskContext.swapCount).toBe(1);
+    });
+
+    it('should limit swap retries to prevent infinite loops', () => {
+      const MAX_SWAPS = 2;
+      let swapCount = 0;
+
+      // Simulate multiple rate limits
+      for (let i = 0; i < 5; i++) {
+        if (swapCount >= MAX_SWAPS) {
+          break; // Should stop after 2 swaps
+        }
+        swapCount++;
+      }
+
+      expect(swapCount).toBe(MAX_SWAPS);
+    });
+  });
+
+  describe('Event Emission Verification', () => {
+    it('should emit sdk-rate-limit event on rate limit', () => {
+      const emitter = new EventEmitter();
+      const sdkRateLimitHandler = vi.fn();
+
+      emitter.on('sdk-rate-limit', sdkRateLimitHandler);
+
+      // Simulate rate limit detected with auto-swap
+      const rateLimitInfo = {
+        source: 'task' as const,
+        taskId: 'task-123',
+        resetTime: 'Dec 17 at 6am',
+        limitType: 'weekly' as const,
+        profileId: 'profile-mai',
+        profileName: 'MAI',
+        wasAutoSwapped: true,
+        swappedToProfile: { id: 'profile-mu', name: 'MU' },
+        swapReason: 'reactive' as const,
+        detectedAt: new Date()
+      };
+
+      emitter.emit('sdk-rate-limit', rateLimitInfo);
+
+      expect(sdkRateLimitHandler).toHaveBeenCalledWith(rateLimitInfo);
+      expect(sdkRateLimitHandler).toHaveBeenCalledTimes(1);
+    });
+
+    it('should emit auto-swap-restart-task event for task restart', () => {
+      const emitter = new EventEmitter();
+      const restartHandler = vi.fn();
+
+      emitter.on('auto-swap-restart-task', restartHandler);
+
+      emitter.emit('auto-swap-restart-task', 'task-123', 'profile-mu');
+
+      expect(restartHandler).toHaveBeenCalledWith('task-123', 'profile-mu');
+    });
+
+    it('should handle event chain: rate-limit → swap → restart', () => {
+      const emitter = new EventEmitter();
+      const events: string[] = [];
+
+      emitter.on('sdk-rate-limit', () => events.push('sdk-rate-limit'));
+      emitter.on('auto-swap-restart-task', () => events.push('auto-swap-restart-task'));
+
+      // Simulate the flow
+      emitter.emit('sdk-rate-limit', { /* info */ });
+      emitter.emit('auto-swap-restart-task', 'task-123', 'profile-mu');
+
+      expect(events).toEqual(['sdk-rate-limit', 'auto-swap-restart-task']);
+    });
+  });
+});
+
+describe('Rate Limit Edge Cases', () => {
+  beforeEach(() => {
+    vi.resetModules();
+  });
+
+  afterEach(() => {
+    vi.clearAllMocks();
+  });
+
+  describe('Pattern Matching Edge Cases', () => {
+    const mockManager = createMockProfileManager();
+
+    beforeEach(() => {
+      vi.doMock('../claude-profile-manager', () => ({
+        getClaudeProfileManager: vi.fn(() => mockManager)
+      }));
+    });
+
+    it('should handle different bullet characters', async () => {
+      const { detectRateLimit } = await import('../rate-limit-detector');
+
+      // Middle dot (·)
+      expect(detectRateLimit('Limit reached · resets 5pm').isRateLimited).toBe(true);
+
+      // Bullet (•)
+      expect(detectRateLimit('Limit reached • resets 5pm').isRateLimited).toBe(true);
+    });
+
+    it('should handle different timezone formats', async () => {
+      const { detectRateLimit } = await import('../rate-limit-detector');
+
+      const timezones = [
+        'Dec 17 at 6am (Europe/Oslo)',
+        'Dec 17 at 6am (America/New_York)',
+        'Dec 17 at 6am (Asia/Tokyo)',
+        'Dec 17 at 6am (UTC)',
+        'Dec 17 at 6am' // No timezone
+      ];
+
+      for (const tz of timezones) {
+        const result = detectRateLimit(`Limit reached · resets ${tz}`);
+        expect(result.isRateLimited).toBe(true);
+      }
+    });
+
+    it('should handle 12-hour and 24-hour time formats', async () => {
+      const { detectRateLimit } = await import('../rate-limit-detector');
+
+      expect(detectRateLimit('Limit reached · resets 11:59pm').isRateLimited).toBe(true);
+      expect(detectRateLimit('Limit reached · resets 6am').isRateLimited).toBe(true);
+      expect(detectRateLimit('Limit reached · resets 18:00').isRateLimited).toBe(true);
+    });
+
+    it('should NOT false-positive on similar messages', async () => {
+      const { detectRateLimit } = await import('../rate-limit-detector');
+
+      // These should NOT trigger rate limit detection
+      const falsePositives = [
+        'Limit your requests to avoid issues', // Contains 'limit' but not rate limit
+        'The speed limit is 60mph', // Unrelated limit
+        'Character limit reached for input field' // Different kind of limit
+      ];
+
+      for (const msg of falsePositives) {
+        const result = detectRateLimit(msg);
+        // Note: Some may still match secondary indicators - that's intentional
+        // The primary pattern should NOT match these
+        const primaryPattern = /Limit reached\s*[·•]\s*resets/i;
+        expect(primaryPattern.test(msg)).toBe(false);
+      }
+    });
+  });
+
+  describe('Both Profiles Rate Limited', () => {
+    it('should return null when all profiles are rate limited', () => {
+      const bothLimitedManager = createMockProfileManager({
+        bestAvailableProfile: null
+      });
+
+      const best = bothLimitedManager.getBestAvailableProfile('profile-mai');
+      expect(best).toBeNull();
+    });
+
+    it('should show manual modal when no profiles available', () => {
+      // User must either wait or add a new account
+      const bothLimitedManager = createMockProfileManager({
+        bestAvailableProfile: null
+      });
+
+      const settings = bothLimitedManager.getAutoSwitchSettings();
+      const bestProfile = bothLimitedManager.getBestAvailableProfile('profile-mai');
+
+      // Even with auto-switch enabled, should show modal since no alternative
+      const shouldShowManualModal = settings.enabled && settings.autoSwitchOnRateLimit && !bestProfile;
+      expect(shouldShowManualModal).toBe(true);
+    });
+  });
+
+  describe('Rapid Rate Limit Succession', () => {
+    it('should enforce max swap count', () => {
+      const MAX_SWAP_COUNT = 2;
+      const context = { swapCount: 0 };
+
+      // First swap
+      context.swapCount++;
+      expect(context.swapCount < MAX_SWAP_COUNT).toBe(true);
+
+      // Second swap
+      context.swapCount++;
+      expect(context.swapCount >= MAX_SWAP_COUNT).toBe(true);
+
+      // Third swap should be blocked
+      const shouldAllowSwap = context.swapCount < MAX_SWAP_COUNT;
+      expect(shouldAllowSwap).toBe(false);
+    });
+  });
+});
+
+describe('Modal Behavior with Reactive Recovery', () => {
+  describe('Modal Content Variations', () => {
+    it('should show notification-style modal when auto-swapped', () => {
+      const rateLimitInfo = {
+        source: 'task' as const,
+        wasAutoSwapped: true,
+        swappedToProfile: { id: 'profile-mu', name: 'MU' },
+        swapReason: 'reactive' as const
+      };
+
+      // When wasAutoSwapped is true, modal should be informational
+      expect(rateLimitInfo.wasAutoSwapped).toBe(true);
+      expect(rateLimitInfo.swapReason).toBe('reactive');
+    });
+
+    it('should show action-required modal when NOT auto-swapped', () => {
+      const rateLimitInfo = {
+        source: 'task' as const,
+        wasAutoSwapped: false,
+        suggestedProfile: { id: 'profile-mu', name: 'MU' }
+      };
+
+      // When wasAutoSwapped is false, user needs to take action
+      expect(rateLimitInfo.wasAutoSwapped).toBe(false);
+    });
+
+    it('should distinguish proactive vs reactive swaps', () => {
+      const proactiveSwap = {
+        wasAutoSwapped: true,
+        swapReason: 'proactive' as const // Before limit hit
+      };
+
+      const reactiveSwap = {
+        wasAutoSwapped: true,
+        swapReason: 'reactive' as const // After limit hit
+      };
+
+      expect(proactiveSwap.swapReason).toBe('proactive');
+      expect(reactiveSwap.swapReason).toBe('reactive');
+    });
+  });
+});

auto-claude-ui/src/main/agent/agent-manager.ts

@@ -9,8 +9,9 @@ import { getClaudeProfileManager } from '../claude-profile-manager';
 import {
   SpecCreationMetadata,
   TaskExecutionOptions,
-  IdeationConfig
+  RoadmapConfig
 } from './types';
+import type { IdeationConfig } from '../../shared/types';
 
 /**
  * Main AgentManager - orchestrates agent process lifecycle
@@ -42,8 +43,10 @@ export class AgentManager extends EventEmitter {
     this.queueManager = new AgentQueueManager(this.state, this.events, this.processManager, this);
 
     // Listen for auto-swap restart events
-    this.on('auto-swap-restart-task', (taskId: string, _newProfileId: string) => {
-      this.restartTask(taskId);
+    this.on('auto-swap-restart-task', (taskId: string, newProfileId: string) => {
+      console.log('[AgentManager] Received auto-swap-restart-task event:', { taskId, newProfileId });
+      const success = this.restartTask(taskId, newProfileId);
+      console.log('[AgentManager] Task restart result:', success ? 'SUCCESS' : 'FAILED');
     });
 
     // Listen for task completion to clean up context (prevent memory leak)
@@ -243,9 +246,11 @@ export class AgentManager extends EventEmitter {
     projectId: string,
     projectPath: string,
     refresh: boolean = false,
-    enableCompetitorAnalysis: boolean = false
+    enableCompetitorAnalysis: boolean = false,
+    refreshCompetitorAnalysis: boolean = false,
+    config?: RoadmapConfig
   ): void {
-    this.queueManager.startRoadmapGeneration(projectId, projectPath, refresh, enableCompetitorAnalysis);
+    this.queueManager.startRoadmapGeneration(projectId, projectPath, refresh, enableCompetitorAnalysis, refreshCompetitorAnalysis, config);
   }
 
   /**
@@ -347,28 +352,56 @@ export class AgentManager extends EventEmitter {
 
   /**
    * Restart task after profile swap
+   * @param taskId - The task to restart
+   * @param newProfileId - Optional new profile ID to apply (from auto-swap)
    */
-  restartTask(taskId: string): boolean {
+  restartTask(taskId: string, newProfileId?: string): boolean {
+    console.log('[AgentManager] restartTask called for:', taskId, 'with newProfileId:', newProfileId);
+
     const context = this.taskExecutionContext.get(taskId);
     if (!context) {
       console.error('[AgentManager] No context for task:', taskId);
+      console.log('[AgentManager] Available task contexts:', Array.from(this.taskExecutionContext.keys()));
       return false;
     }
 
+    console.log('[AgentManager] Task context found:', {
+      taskId,
+      projectPath: context.projectPath,
+      specId: context.specId,
+      isSpecCreation: context.isSpecCreation,
+      swapCount: context.swapCount
+    });
+
     // Prevent infinite swap loops
     if (context.swapCount >= 2) {
-      console.error('[AgentManager] Max swap count reached for task:', taskId);
+      console.error('[AgentManager] Max swap count reached for task:', taskId, '- stopping restart loop');
       return false;
     }
 
     context.swapCount++;
+    console.log('[AgentManager] Incremented swap count to:', context.swapCount);
+
+    // If a new profile was specified, ensure it's set as active before restart
+    if (newProfileId) {
+      const profileManager = getClaudeProfileManager();
+      const currentActiveId = profileManager.getActiveProfile()?.id;
+      if (currentActiveId !== newProfileId) {
+        console.log('[AgentManager] Setting active profile to:', newProfileId);
+        profileManager.setActiveProfile(newProfileId);
+      }
+    }
 
     // Kill current process
+    console.log('[AgentManager] Killing current process for task:', taskId);
     this.killTask(taskId);
 
     // Wait for cleanup, then restart
+    console.log('[AgentManager] Scheduling task restart in 500ms');
     setTimeout(() => {
+      console.log('[AgentManager] Restarting task now:', taskId);
       if (context.isSpecCreation) {
+        console.log('[AgentManager] Restarting as spec creation');
         this.startSpecCreation(
           taskId,
           context.projectPath,
@@ -377,6 +410,7 @@ export class AgentManager extends EventEmitter {
           context.metadata
         );
       } else {
+        console.log('[AgentManager] Restarting as task execution');
         this.startTaskExecution(
           taskId,
           context.projectPath,

auto-claude-ui/src/main/agent/agent-process.ts

@@ -9,6 +9,7 @@ import { ProcessType, ExecutionProgressData } from './types';
 import { detectRateLimit, createSDKRateLimitInfo, getProfileEnv, detectAuthFailure } from '../rate-limit-detector';
 import { projectStore } from '../project-store';
 import { getClaudeProfileManager } from '../claude-profile-manager';
+import { findPythonCommand, parsePythonCommand } from '../python-detector';
 
 /**
  * Process spawning and lifecycle management
@@ -17,7 +18,8 @@ export class AgentProcessManager {
   private state: AgentState;
   private events: AgentEvents;
   private emitter: EventEmitter;
-  private pythonPath: string = 'python3';
+  // Auto-detect Python command on initialization
+  private pythonPath: string = findPythonCommand() || 'python';
   private autoBuildSourcePath: string = '';
 
   constructor(state: AgentState, events: AgentEvents, emitter: EventEmitter) {
@@ -161,7 +163,9 @@ export class AgentProcessManager {
     // Get active Claude profile environment (CLAUDE_CONFIG_DIR if not default)
     const profileEnv = getProfileEnv();
 
-    const childProcess = spawn(this.pythonPath, args, {
+    // Parse Python command to handle space-separated commands like "py -3"
+    const [pythonCommand, pythonBaseArgs] = parsePythonCommand(this.pythonPath);
+    const childProcess = spawn(pythonCommand, [...pythonBaseArgs, ...args], {
       cwd,
       env: {
         ...process.env,
@@ -237,6 +241,10 @@ export class AgentProcessManager {
       const log = data.toString('utf8');
       this.emitter.emit('log', taskId, log);
       processLog(log);
+      // Print to console when DEBUG is enabled (visible in pnpm dev terminal)
+      if (['true', '1', 'yes', 'on'].includes(process.env.DEBUG?.toLowerCase() ?? '')) {
+        console.log(`[Agent:${taskId}] ${log.trim()}`);
+      }
     });
 
     // Handle stderr - explicitly decode as UTF-8 for cross-platform Unicode support
@@ -246,6 +254,10 @@ export class AgentProcessManager {
       // so we treat it as log, not error
       this.emitter.emit('log', taskId, log);
       processLog(log);
+      // Print to console when DEBUG is enabled (visible in pnpm dev terminal)
+      if (['true', '1', 'yes', 'on'].includes(process.env.DEBUG?.toLowerCase() ?? '')) {
+        console.log(`[Agent:${taskId}] ${log.trim()}`);
+      }
     });
 
     // Handle process exit
@@ -261,22 +273,45 @@ export class AgentProcessManager {
 
       // Check for rate limit if process failed
       if (code !== 0) {
+        console.log('[AgentProcess] Process failed with code:', code, 'for task:', taskId);
+        console.log('[AgentProcess] Checking for rate limit in output (last 500 chars):', allOutput.slice(-500));
+
         const rateLimitDetection = detectRateLimit(allOutput);
+        console.log('[AgentProcess] Rate limit detection result:', {
+          isRateLimited: rateLimitDetection.isRateLimited,
+          resetTime: rateLimitDetection.resetTime,
+          limitType: rateLimitDetection.limitType,
+          profileId: rateLimitDetection.profileId,
+          suggestedProfile: rateLimitDetection.suggestedProfile
+        });
+
         if (rateLimitDetection.isRateLimited) {
           // Check if auto-swap is enabled
           const profileManager = getClaudeProfileManager();
           const autoSwitchSettings = profileManager.getAutoSwitchSettings();
 
+          console.log('[AgentProcess] Auto-switch settings:', {
+            enabled: autoSwitchSettings.enabled,
+            autoSwitchOnRateLimit: autoSwitchSettings.autoSwitchOnRateLimit,
+            proactiveSwapEnabled: autoSwitchSettings.proactiveSwapEnabled
+          });
+
           if (autoSwitchSettings.enabled && autoSwitchSettings.autoSwitchOnRateLimit) {
             const currentProfileId = rateLimitDetection.profileId;
             const bestProfile = profileManager.getBestAvailableProfile(currentProfileId);
 
+            console.log('[AgentProcess] Best available profile:', bestProfile ? {
+              id: bestProfile.id,
+              name: bestProfile.name
+            } : 'NONE');
+
             if (bestProfile) {
               // Switch active profile
+              console.log('[AgentProcess] AUTO-SWAP: Switching from', currentProfileId, 'to', bestProfile.id);
               profileManager.setActiveProfile(bestProfile.id);
 
               // Emit swap info (for modal)
-              const source = processType === 'spec-creation' ? 'task' : 'task';
+              const source = processType === 'spec-creation' ? 'roadmap' : 'task';
               const rateLimitInfo = createSDKRateLimitInfo(source, rateLimitDetection, {
                 taskId
               });
@@ -286,30 +321,42 @@ export class AgentProcessManager {
                 name: bestProfile.name
               };
               rateLimitInfo.swapReason = 'reactive';
+
+              console.log('[AgentProcess] Emitting sdk-rate-limit event (auto-swapped):', rateLimitInfo);
               this.emitter.emit('sdk-rate-limit', rateLimitInfo);
 
               // Restart task
+              console.log('[AgentProcess] Emitting auto-swap-restart-task event for task:', taskId);
               this.emitter.emit('auto-swap-restart-task', taskId, bestProfile.id);
               return;
+            } else {
+              console.log('[AgentProcess] No alternative profile available - falling back to manual modal');
             }
+          } else {
+            console.log('[AgentProcess] Auto-switch disabled - showing manual modal');
           }
 
           // Fall back to manual modal (no auto-swap or no alternative profile)
-          const source = processType === 'spec-creation' ? 'task' : 'task';
+          const source = processType === 'spec-creation' ? 'roadmap' : 'task';
           const rateLimitInfo = createSDKRateLimitInfo(source, rateLimitDetection, {
             taskId
           });
+          console.log('[AgentProcess] Emitting sdk-rate-limit event (manual):', rateLimitInfo);
           this.emitter.emit('sdk-rate-limit', rateLimitInfo);
         } else {
+          console.log('[AgentProcess] No rate limit detected - checking for auth failure');
           // Not rate limited - check for authentication failure
           const authFailureDetection = detectAuthFailure(allOutput);
           if (authFailureDetection.isAuthFailure) {
+            console.log('[AgentProcess] Auth failure detected:', authFailureDetection);
             this.emitter.emit('auth-failure', taskId, {
               profileId: authFailureDetection.profileId,
               failureType: authFailureDetection.failureType,
               message: authFailureDetection.message,
               originalError: authFailureDetection.originalError
             });
+          } else {
+            console.log('[AgentProcess] Process failed but no rate limit or auth failure detected');
           }
         }
       }

auto-claude-ui/src/main/agent/agent-queue.ts

@@ -5,9 +5,12 @@ import { EventEmitter } from 'events';
 import { AgentState } from './agent-state';
 import { AgentEvents } from './agent-events';
 import { AgentProcessManager } from './agent-process';
-import { IdeationConfig } from './types';
+import { RoadmapConfig } from './types';
+import type { IdeationConfig } from '../../shared/types';
+import { MODEL_ID_MAP } from '../../shared/constants';
 import { detectRateLimit, createSDKRateLimitInfo, getProfileEnv } from '../rate-limit-detector';
 import { debugLog, debugError } from '../../shared/utils/debug-logger';
+import { parsePythonCommand } from '../python-detector';
 
 /**
  * Queue management for ideation and roadmap generation
@@ -32,18 +35,26 @@ export class AgentQueueManager {
 
   /**
    * Start roadmap generation process
+   *
+   * @param refreshCompetitorAnalysis - Force refresh competitor analysis even if it exists.
+   *   This allows refreshing competitor data independently of the general roadmap refresh.
+   *   Use when user explicitly wants new competitor research.
    */
   startRoadmapGeneration(
     projectId: string,
     projectPath: string,
     refresh: boolean = false,
-    enableCompetitorAnalysis: boolean = false
+    enableCompetitorAnalysis: boolean = false,
+    refreshCompetitorAnalysis: boolean = false,
+    config?: RoadmapConfig
   ): void {
     debugLog('[Agent Queue] Starting roadmap generation:', {
       projectId,
       projectPath,
       refresh,
-      enableCompetitorAnalysis
+      enableCompetitorAnalysis,
+      refreshCompetitorAnalysis,
+      config
     });
 
     const autoBuildSource = this.processManager.getAutoBuildSourcePath();
@@ -73,6 +84,20 @@ export class AgentQueueManager {
       args.push('--competitor-analysis');
     }
 
+    // Add refresh competitor analysis flag if user wants fresh competitor data
+    if (refreshCompetitorAnalysis) {
+      args.push('--refresh-competitor-analysis');
+    }
+
+    // Add model and thinking level from config
+    if (config?.model) {
+      const modelId = MODEL_ID_MAP[config.model] || MODEL_ID_MAP['opus'];
+      args.push('--model', modelId);
+    }
+    if (config?.thinkingLevel) {
+      args.push('--thinking-level', config.thinkingLevel);
+    }
+
     debugLog('[Agent Queue] Spawning roadmap process with args:', args);
 
     // Use projectId as taskId for roadmap operations
@@ -140,6 +165,15 @@ export class AgentQueueManager {
       args.push('--append');
     }
 
+    // Add model and thinking level from config
+    if (config.model) {
+      const modelId = MODEL_ID_MAP[config.model] || MODEL_ID_MAP['opus'];
+      args.push('--model', modelId);
+    }
+    if (config.thinkingLevel) {
+      args.push('--thinking-level', config.thinkingLevel);
+    }
+
     debugLog('[Agent Queue] Spawning ideation process with args:', args);
 
     // Use projectId as taskId for ideation operations
@@ -206,7 +240,9 @@ export class AgentQueueManager {
       tokenPreview: hasToken ? oauthToken?.substring(0, 20) + '...' : 'none'
     });
 
-    const childProcess = spawn(pythonPath, args, {
+    // Parse Python command to handle space-separated commands like "py -3"
+    const [pythonCommand, pythonBaseArgs] = parsePythonCommand(pythonPath);
+    const childProcess = spawn(pythonCommand, [...pythonBaseArgs, ...args], {
       cwd,
       env: finalEnv
     });
@@ -315,6 +351,15 @@ export class AgentQueueManager {
     childProcess.on('exit', (code: number | null) => {
       debugLog('[Agent Queue] Ideation process exited:', { projectId, code, spawnId });
 
+      // Check if this process was intentionally stopped by the user
+      const wasIntentionallyStopped = this.state.wasSpawnKilled(spawnId);
+      if (wasIntentionallyStopped) {
+        debugLog('[Agent Queue] Ideation process was intentionally stopped, ignoring exit');
+        this.state.clearKilledSpawn(spawnId);
+        this.state.deleteProcess(projectId);
+        return;
+      }
+
       // Get the stored project path before deleting from map
       const processInfo = this.state.getProcess(projectId);
       const storedProjectPath = processInfo?.projectPath;
@@ -441,7 +486,9 @@ export class AgentQueueManager {
       tokenPreview: hasToken ? oauthToken?.substring(0, 20) + '...' : 'none'
     });
 
-    const childProcess = spawn(pythonPath, args, {
+    // Parse Python command to handle space-separated commands like "py -3"
+    const [pythonCommand, pythonBaseArgs] = parsePythonCommand(pythonPath);
+    const childProcess = spawn(pythonCommand, [...pythonBaseArgs, ...args], {
       cwd,
       env: finalEnv
     });
@@ -512,6 +559,15 @@ export class AgentQueueManager {
     childProcess.on('exit', (code: number | null) => {
       debugLog('[Agent Queue] Roadmap process exited:', { projectId, code, spawnId });
 
+      // Check if this process was intentionally stopped by the user
+      const wasIntentionallyStopped = this.state.wasSpawnKilled(spawnId);
+      if (wasIntentionallyStopped) {
+        debugLog('[Agent Queue] Roadmap process was intentionally stopped, ignoring exit');
+        this.state.clearKilledSpawn(spawnId);
+        this.state.deleteProcess(projectId);
+        return;
+      }
+
       // Get the stored project path before deleting from map
       const processInfo = this.state.getProcess(projectId);
       const storedProjectPath = processInfo?.projectPath;

auto-claude-ui/src/main/agent/index.ts

@@ -20,9 +20,11 @@ export type {
   ExecutionProgressData,
   ProcessType,
   AgentManagerEvents,
-  IdeationConfig,
   TaskExecutionOptions,
   SpecCreationMetadata,
   IdeationProgressData,
   RoadmapProgressData
 } from './types';
+
+// Re-export IdeationConfig from shared types for consistency
+export type { IdeationConfig } from '../../shared/types';

auto-claude-ui/src/main/agent/types.ts

@@ -1,4 +1,5 @@
 import { ChildProcess } from 'child_process';
+import type { IdeationConfig } from '../../shared/types';
 
 /**
  * Agent-specific types for process and state management
@@ -32,12 +33,11 @@ export interface AgentManagerEvents {
   'execution-progress': (taskId: string, progress: ExecutionProgressData) => void;
 }
 
-export interface IdeationConfig {
-  enabledTypes: string[];
-  includeRoadmapContext: boolean;
-  includeKanbanContext: boolean;
-  maxIdeasPerType: number;
-  append?: boolean;
+// IdeationConfig now imported from shared types to maintain consistency
+
+export interface RoadmapConfig {
+  model?: string;          // Model shorthand (opus, sonnet, haiku)
+  thinkingLevel?: string;  // Thinking level (none, low, medium, high, ultrathink)
 }
 
 export interface TaskExecutionOptions {

auto-claude-ui/src/main/api-validation-service.ts

@@ -0,0 +1,243 @@
+/**
+ * API Validation Service
+ *
+ * Provides validation for external LLM API providers (OpenAI, Anthropic, Google, etc.)
+ * Used by the Graphiti memory integration for embedding and LLM operations.
+ */
+
+export interface ApiValidationResult {
+  success: boolean;
+  message: string;
+  details?: {
+    provider?: string;
+    model?: string;
+    latencyMs?: number;
+  };
+}
+
+/**
+ * Validate OpenAI API key by attempting to list models
+ * @param apiKey - OpenAI API key
+ */
+export async function validateOpenAIApiKey(
+  apiKey: string
+): Promise<ApiValidationResult> {
+  if (!apiKey || !apiKey.trim()) {
+    return {
+      success: false,
+      message: 'API key is required',
+    };
+  }
+
+  // Basic format validation
+  const trimmedKey = apiKey.trim();
+  if (!trimmedKey.startsWith('sk-') && !trimmedKey.startsWith('sess-')) {
+    return {
+      success: false,
+      message: 'Invalid API key format. OpenAI API keys should start with "sk-" or "sess-"',
+    };
+  }
+
+  try {
+    const startTime = Date.now();
+
+    // Use native https module to avoid additional dependencies
+    const result = await new Promise<ApiValidationResult>((resolve) => {
+      const https = require('https');
+
+      const options = {
+        hostname: 'api.openai.com',
+        port: 443,
+        path: '/v1/models',
+        method: 'GET',
+        headers: {
+          Authorization: `Bearer ${trimmedKey}`,
+          'Content-Type': 'application/json',
+        },
+        timeout: 15000,
+      };
+
+      const req = https.request(options, (res: { statusCode: number; on: (event: string, callback: (chunk: Buffer) => void) => void }) => {
+        let data = '';
+
+        res.on('data', (chunk: Buffer) => {
+          data += chunk;
+        });
+
+        res.on('end', () => {
+          const latencyMs = Date.now() - startTime;
+
+          if (res.statusCode === 200) {
+            resolve({
+              success: true,
+              message: 'OpenAI API key is valid',
+              details: {
+                provider: 'openai',
+                latencyMs,
+              },
+            });
+          } else if (res.statusCode === 401) {
+            resolve({
+              success: false,
+              message: 'Invalid API key. Please check your OpenAI API key.',
+            });
+          } else if (res.statusCode === 429) {
+            // Rate limited but key is valid
+            resolve({
+              success: true,
+              message: 'OpenAI API key is valid (rate limited, please wait)',
+              details: {
+                provider: 'openai',
+                latencyMs,
+              },
+            });
+          } else {
+            try {
+              const errorData = JSON.parse(data);
+              resolve({
+                success: false,
+                message: errorData.error?.message || `API error: ${res.statusCode}`,
+              });
+            } catch {
+              resolve({
+                success: false,
+                message: `API error: ${res.statusCode}`,
+              });
+            }
+          }
+        });
+      });
+
+      req.on('error', (error: Error) => {
+        resolve({
+          success: false,
+          message: `Connection error: ${error.message}`,
+        });
+      });
+
+      req.on('timeout', () => {
+        req.destroy();
+        resolve({
+          success: false,
+          message: 'Connection timeout. Please check your network connection.',
+        });
+      });
+
+      req.end();
+    });
+
+    return result;
+  } catch (error) {
+    return {
+      success: false,
+      message: error instanceof Error ? error.message : 'Unknown error occurred',
+    };
+  }
+}
+
+/**
+ * Validate Anthropic API key
+ * @param apiKey - Anthropic API key
+ */
+export async function validateAnthropicApiKey(
+  apiKey: string
+): Promise<ApiValidationResult> {
+  if (!apiKey || !apiKey.trim()) {
+    return {
+      success: false,
+      message: 'API key is required',
+    };
+  }
+
+  const trimmedKey = apiKey.trim();
+  if (!trimmedKey.startsWith('sk-ant-')) {
+    return {
+      success: false,
+      message: 'Invalid API key format. Anthropic API keys should start with "sk-ant-"',
+    };
+  }
+
+  // For now, just validate format - full validation would require an API call
+  return {
+    success: true,
+    message: 'Anthropic API key format is valid',
+    details: {
+      provider: 'anthropic',
+    },
+  };
+}
+
+/**
+ * Validate Google AI API key
+ * @param apiKey - Google AI API key
+ */
+export async function validateGoogleApiKey(
+  apiKey: string
+): Promise<ApiValidationResult> {
+  if (!apiKey || !apiKey.trim()) {
+    return {
+      success: false,
+      message: 'API key is required',
+    };
+  }
+
+  const trimmedKey = apiKey.trim();
+  if (!trimmedKey.startsWith('AIza')) {
+    return {
+      success: false,
+      message: 'Invalid API key format. Google AI API keys should start with "AIza"',
+    };
+  }
+
+  return {
+    success: true,
+    message: 'Google AI API key format is valid',
+    details: {
+      provider: 'google',
+    },
+  };
+}
+
+/**
+ * Validate an LLM provider API key based on provider type
+ * @param provider - The LLM provider (openai, anthropic, google, etc.)
+ * @param apiKey - The API key to validate
+ */
+export async function validateLLMApiKey(
+  provider: string,
+  apiKey: string
+): Promise<ApiValidationResult> {
+  switch (provider) {
+    case 'openai':
+      return validateOpenAIApiKey(apiKey);
+    case 'anthropic':
+      return validateAnthropicApiKey(apiKey);
+    case 'google':
+      return validateGoogleApiKey(apiKey);
+    case 'ollama':
+      // Ollama is local, no API key needed
+      return {
+        success: true,
+        message: 'Ollama runs locally, no API key required',
+        details: { provider: 'ollama' },
+      };
+    case 'azure_openai':
+      // Azure OpenAI uses different auth, just validate presence
+      if (!apiKey || !apiKey.trim()) {
+        return {
+          success: false,
+          message: 'Azure OpenAI API key is required',
+        };
+      }
+      return {
+        success: true,
+        message: 'Azure OpenAI API key format accepted',
+        details: { provider: 'azure_openai' },
+      };
+    default:
+      return {
+        success: false,
+        message: `Unknown provider: ${provider}`,
+      };
+  }
+}

auto-claude-ui/src/main/app-updater.ts

@@ -23,8 +23,8 @@ import type { BrowserWindow } from 'electron';
 import { IPC_CHANNELS } from '../shared/constants';
 import type { AppUpdateInfo } from '../shared/types';
 
-// Debug mode - set via environment variable
-const DEBUG_UPDATER = process.env.DEBUG_UPDATER === 'true' || process.env.DEBUG === 'true';
+// Debug mode - DEBUG_UPDATER=true or development mode
+const DEBUG_UPDATER = process.env.DEBUG_UPDATER === 'true' || process.env.NODE_ENV === 'development';
 
 // Configure electron-updater
 autoUpdater.autoDownload = true;  // Automatically download updates when available

auto-claude-ui/src/main/auto-claude-updater.ts

@@ -27,7 +27,7 @@ export type {
 } from './updater/types';
 
 // Export version management
-export { getBundledVersion } from './updater/version-manager';
+export { getBundledVersion, getEffectiveVersion } from './updater/version-manager';
 
 // Export path resolution
 export {

auto-claude-ui/src/main/changelog/changelog-service.ts

@@ -27,13 +27,15 @@ import {
   getCommits,
   getBranchDiffCommits
 } from './git-integration';
+import { findPythonCommand } from '../python-detector';
 
 /**
  * Main changelog service - orchestrates all changelog operations
  * Delegates to specialized modules for specific concerns
  */
 export class ChangelogService extends EventEmitter {
-  private pythonPath: string = 'python3';
+  // Auto-detect Python command on initialization
+  private pythonPath: string = findPythonCommand() || 'python';
   private claudePath: string = 'claude';
   private autoBuildSourcePath: string = '';
   private cachedEnv: Record<string, string> | null = null;
@@ -89,7 +91,7 @@ export class ChangelogService extends EventEmitter {
 
   /**
    * Check if debug mode is enabled
-   * Checks DEBUG from auto-claude/.env and AUTO_CLAUDE_DEBUG from process.env
+   * Checks DEBUG from auto-claude/.env and DEBUG from process.env
    */
   private isDebugEnabled(): boolean {
     // Cache the result after first check
@@ -101,8 +103,8 @@ export class ChangelogService extends EventEmitter {
     if (
       process.env.DEBUG === 'true' ||
       process.env.DEBUG === '1' ||
-      process.env.AUTO_CLAUDE_DEBUG === 'true' ||
-      process.env.AUTO_CLAUDE_DEBUG === '1'
+      process.env.DEBUG === 'true' ||
+      process.env.DEBUG === '1'
     ) {
       this.debugEnabled = true;
       return true;
@@ -115,7 +117,7 @@ export class ChangelogService extends EventEmitter {
   }
 
   /**
-   * Debug logging - only logs when DEBUG=true in auto-claude/.env or AUTO_CLAUDE_DEBUG is set
+   * Debug logging - only logs when DEBUG=true in auto-claude/.env or DEBUG is set
    */
   private debug(...args: unknown[]): void {
     if (this.isDebugEnabled()) {

auto-claude-ui/src/main/changelog/formatter.ts

@@ -49,7 +49,11 @@ const FORMAT_TEMPLATES = {
 
 ## What's Changed
 
-- type: description by @contributor in commit-hash`
+- type: description by @contributor in commit-hash
+
+## Thanks to all contributors
+
+@contributor1, @contributor2`
 };
 
 /**
@@ -274,7 +278,15 @@ PART 2 - "What's Changed" (raw commit list):
 - Example: "- feat: add dark mode support by @contributor in def5678"
 - Include the commit type prefix (feat:, fix:, docs:, etc.)
 - Show the author name with @ prefix
-- Show the short commit hash at the end`;
+- Show the short commit hash at the end
+
+PART 3 - "Thanks to all contributors" (deduplicated list):
+- Add this section after "What's Changed"
+- Extract all unique contributor names from the commits
+- List them in a comma-separated format with @ prefix
+- Example: "## Thanks to all contributors\\n\\n@contributor1, @contributor2, @contributor3"
+- Only include unique names (no duplicates)
+- This acknowledges everyone who contributed to this release`;
   }
 
   return `${audienceInstruction}

auto-claude-ui/src/main/changelog/generator.ts

@@ -12,6 +12,7 @@ import { buildChangelogPrompt, buildGitPrompt, createGenerationScript } from './
 import { extractChangelog } from './parser';
 import { getCommits, getBranchDiffCommits } from './git-integration';
 import { detectRateLimit, createSDKRateLimitInfo, getProfileEnv } from '../rate-limit-detector';
+import { parsePythonCommand } from '../python-detector';
 
 /**
  * Core changelog generation logic
@@ -139,7 +140,9 @@ export class ChangelogGenerator extends EventEmitter {
     // Build environment with explicit critical variables
     const spawnEnv = this.buildSpawnEnvironment();
 
-    const childProcess = spawn(this.pythonPath, ['-c', script], {
+    // Parse Python command to handle space-separated commands like "py -3"
+    const [pythonCommand, pythonBaseArgs] = parsePythonCommand(this.pythonPath);
+    const childProcess = spawn(pythonCommand, [...pythonBaseArgs, '-c', script], {
       cwd: this.autoBuildSourcePath,
       env: spawnEnv
     });

auto-claude-ui/src/main/changelog/version-suggester.ts

@@ -3,6 +3,7 @@ import * as path from 'path';
 import * as os from 'os';
 import type { GitCommit } from '../../shared/types';
 import { getProfileEnv } from '../rate-limit-detector';
+import { parsePythonCommand } from '../python-detector';
 
 interface VersionSuggestion {
   version: string;
@@ -52,7 +53,9 @@ export class VersionSuggester {
     const spawnEnv = this.buildSpawnEnvironment();
 
     return new Promise((resolve, _reject) => {
-      const childProcess = spawn(this.pythonPath, ['-c', script], {
+      // Parse Python command to handle space-separated commands like "py -3"
+      const [pythonCommand, pythonBaseArgs] = parsePythonCommand(this.pythonPath);
+      const childProcess = spawn(pythonCommand, [...pythonBaseArgs, '-c', script], {
         cwd: this.autoBuildSourcePath,
         env: spawnEnv
       });

auto-claude-ui/src/main/claude-profile/usage-monitor.ts

@@ -99,7 +99,9 @@ export class UsageMonitor extends EventEmitter {
       }
 
       // Fetch current usage (hybrid approach)
-      const usage = await this.fetchUsage(activeProfile.id, activeProfile.oauthToken);
+      // Get decrypted token from ProfileManager (activeProfile.oauthToken is encrypted)
+      const decryptedToken = profileManager.getProfileToken(activeProfile.id);
+      const usage = await this.fetchUsage(activeProfile.id, decryptedToken ?? undefined);
       if (!usage) {
         console.warn('[UsageMonitor] Failed to fetch usage');
         return;

auto-claude-ui/src/main/docker-service.ts

@@ -1,586 +0,0 @@
-/**
- * Docker & FalkorDB Service
- *
- * Provides automatic detection and management of Docker and FalkorDB
- * for non-technical users. This eliminates the need for manual
- * "docker --version" verification steps.
- */
-
-import { exec, spawn } from 'child_process';
-import { promisify } from 'util';
-
-const execAsync = promisify(exec);
-
-// FalkorDB container configuration
-const FALKORDB_CONTAINER_NAME = 'auto-claude-falkordb';
-const FALKORDB_IMAGE = 'falkordb/falkordb:latest';
-const FALKORDB_DEFAULT_PORT = 6380;
-
-export interface DockerStatus {
-  installed: boolean;
-  running: boolean;
-  version?: string;
-  error?: string;
-}
-
-export interface FalkorDBStatus {
-  containerExists: boolean;
-  containerRunning: boolean;
-  containerName: string;
-  port: number;
-  healthy: boolean;
-  error?: string;
-}
-
-export interface InfrastructureStatus {
-  docker: DockerStatus;
-  falkordb: FalkorDBStatus;
-  ready: boolean; // True if both Docker is running and FalkorDB is healthy
-}
-
-/**
- * Check if Docker is installed and running
- */
-export async function checkDockerStatus(): Promise<DockerStatus> {
-  try {
-    // Check if Docker CLI is available
-    const { stdout: versionOutput } = await execAsync('docker --version', {
-      timeout: 5000,
-    });
-
-    const version = versionOutput.trim();
-
-    // Check if Docker daemon is running by trying to ping it
-    try {
-      await execAsync('docker info', { timeout: 10000 });
-      return {
-        installed: true,
-        running: true,
-        version,
-      };
-    } catch {
-      return {
-        installed: true,
-        running: false,
-        version,
-        error: 'Docker is installed but not running. Please start Docker Desktop.',
-      };
-    }
-  } catch (error) {
-    const errorMsg = error instanceof Error ? error.message : String(error);
-
-    // Check if it's a "command not found" type error
-    if (
-      errorMsg.includes('not found') ||
-      errorMsg.includes('ENOENT') ||
-      errorMsg.includes('not recognized')
-    ) {
-      return {
-        installed: false,
-        running: false,
-        error: 'Docker is not installed. Please install Docker Desktop.',
-      };
-    }
-
-    return {
-      installed: false,
-      running: false,
-      error: `Docker check failed: ${errorMsg}`,
-    };
-  }
-}
-
-/**
- * Get the actual port mapping for the FalkorDB container from Docker
- */
-async function getContainerPortMapping(): Promise<number | null> {
-  try {
-    // Get the port mapping from Docker - format: "0.0.0.0:6380->6379/tcp"
-    const { stdout } = await execAsync(
-      `docker port ${FALKORDB_CONTAINER_NAME} 6379`,
-      { timeout: 5000 }
-    );
-
-    const portMatch = stdout.trim().match(/:(\d+)/);
-    if (portMatch) {
-      return parseInt(portMatch[1], 10);
-    }
-    return null;
-  } catch {
-    return null;
-  }
-}
-
-/**
- * Check FalkorDB container status
- */
-export async function checkFalkorDBStatus(port: number = FALKORDB_DEFAULT_PORT): Promise<FalkorDBStatus> {
-  const status: FalkorDBStatus = {
-    containerExists: false,
-    containerRunning: false,
-    containerName: FALKORDB_CONTAINER_NAME,
-    port,
-    healthy: false,
-  };
-
-  try {
-    // Check if container exists and get its status
-    const { stdout } = await execAsync(
-      `docker ps -a --filter "name=${FALKORDB_CONTAINER_NAME}" --format "{{.Status}}"`,
-      { timeout: 5000 }
-    );
-
-    const containerStatus = stdout.trim();
-
-    if (containerStatus) {
-      status.containerExists = true;
-      status.containerRunning = containerStatus.toLowerCase().startsWith('up');
-
-      if (status.containerRunning) {
-        // Get the actual port mapping from Docker
-        const actualPort = await getContainerPortMapping();
-        if (actualPort) {
-          status.port = actualPort;
-        }
-
-        // Check if FalkorDB is responding
-        status.healthy = await checkFalkorDBHealth(status.port);
-      }
-    }
-
-    return status;
-  } catch (error) {
-    status.error = error instanceof Error ? error.message : String(error);
-    return status;
-  }
-}
-
-/**
- * Check if FalkorDB is responding to connections
- */
-async function checkFalkorDBHealth(_port: number): Promise<boolean> {
-  try {
-    // Try to ping FalkorDB using redis-cli (FalkorDB uses Redis protocol)
-    // Since we may not have redis-cli, we'll check if the port is listening
-    await execAsync(`docker exec ${FALKORDB_CONTAINER_NAME} redis-cli PING`, {
-      timeout: 5000,
-    });
-    return true;
-  } catch {
-    // Fallback: just check if container is running (less accurate)
-    return false;
-  }
-}
-
-/**
- * Get combined infrastructure status
- */
-export async function getInfrastructureStatus(
-  falkordbPort: number = FALKORDB_DEFAULT_PORT
-): Promise<InfrastructureStatus> {
-  const [docker, falkordb] = await Promise.all([
-    checkDockerStatus(),
-    checkFalkorDBStatus(falkordbPort),
-  ]);
-
-  return {
-    docker,
-    falkordb,
-    ready: docker.running && falkordb.containerRunning && falkordb.healthy,
-  };
-}
-
-/**
- * Start FalkorDB container
- * Creates a new container if it doesn't exist, or starts the existing one
- */
-export async function startFalkorDB(
-  port: number = FALKORDB_DEFAULT_PORT
-): Promise<{ success: boolean; error?: string }> {
-  try {
-    // First, check Docker status
-    const dockerStatus = await checkDockerStatus();
-    if (!dockerStatus.running) {
-      return {
-        success: false,
-        error: dockerStatus.error || 'Docker is not running',
-      };
-    }
-
-    // Check if container already exists
-    const falkordbStatus = await checkFalkorDBStatus(port);
-
-    if (falkordbStatus.containerExists) {
-      if (falkordbStatus.containerRunning) {
-        // Already running
-        return { success: true };
-      }
-
-      // Start existing container
-      await execAsync(`docker start ${FALKORDB_CONTAINER_NAME}`, { timeout: 30000 });
-    } else {
-      // Create and start new container
-      await execAsync(
-        `docker run -d --name ${FALKORDB_CONTAINER_NAME} -p ${port}:6379 ${FALKORDB_IMAGE}`,
-        { timeout: 60000 }
-      );
-    }
-
-    // Wait for FalkorDB to be ready (up to 30 seconds)
-    const ready = await waitForFalkorDB(port, 30000);
-
-    if (!ready) {
-      return {
-        success: false,
-        error: 'FalkorDB container started but is not responding. Please check Docker logs.',
-      };
-    }
-
-    return { success: true };
-  } catch (error) {
-    return {
-      success: false,
-      error: error instanceof Error ? error.message : String(error),
-    };
-  }
-}
-
-/**
- * Stop FalkorDB container
- */
-export async function stopFalkorDB(): Promise<{ success: boolean; error?: string }> {
-  try {
-    await execAsync(`docker stop ${FALKORDB_CONTAINER_NAME}`, { timeout: 30000 });
-    return { success: true };
-  } catch (error) {
-    return {
-      success: false,
-      error: error instanceof Error ? error.message : String(error),
-    };
-  }
-}
-
-/**
- * Wait for FalkorDB to be ready
- */
-async function waitForFalkorDB(port: number, timeoutMs: number): Promise<boolean> {
-  const startTime = Date.now();
-  const checkInterval = 1000; // Check every second
-
-  while (Date.now() - startTime < timeoutMs) {
-    const status = await checkFalkorDBStatus(port);
-    if (status.containerRunning && status.healthy) {
-      return true;
-    }
-    // If container is running but not healthy yet, wait
-    if (status.containerRunning) {
-      await new Promise((resolve) => setTimeout(resolve, checkInterval));
-    } else {
-      // Container stopped unexpectedly
-      return false;
-    }
-  }
-
-  return false;
-}
-
-/**
- * Open Docker Desktop application (macOS/Windows)
- */
-export async function openDockerDesktop(): Promise<{ success: boolean; error?: string }> {
-  try {
-    if (process.platform === 'darwin') {
-      // macOS
-      await execAsync('open -a Docker', { timeout: 5000 });
-    } else if (process.platform === 'win32') {
-      // Windows
-      spawn('cmd', ['/c', 'start', '', 'Docker Desktop'], {
-        detached: true,
-        stdio: 'ignore',
-      });
-    } else {
-      // Linux - Docker doesn't have a GUI, suggest starting daemon
-      return {
-        success: false,
-        error: 'On Linux, start Docker with: sudo systemctl start docker',
-      };
-    }
-
-    return { success: true };
-  } catch (error) {
-    return {
-      success: false,
-      error: error instanceof Error ? error.message : String(error),
-    };
-  }
-}
-
-/**
- * Get download URL for Docker Desktop
- */
-export function getDockerDownloadUrl(): string {
-  if (process.platform === 'darwin') {
-    return 'https://www.docker.com/products/docker-desktop/';
-  } else if (process.platform === 'win32') {
-    return 'https://www.docker.com/products/docker-desktop/';
-  }
-  return 'https://docs.docker.com/engine/install/';
-}
-
-// ============================================
-// Graphiti Validation Functions
-// ============================================
-
-export interface GraphitiValidationResult {
-  success: boolean;
-  message: string;
-  details?: {
-    provider?: string;
-    model?: string;
-    latencyMs?: number;
-  };
-}
-
-/**
- * Validate FalkorDB connection by attempting to connect and ping
- * @param uri - FalkorDB URI (e.g., "bolt://localhost:6380" or "redis://localhost:6380")
- */
-export async function validateFalkorDBConnection(
-  uri: string
-): Promise<GraphitiValidationResult> {
-  try {
-    // Parse the URI to extract host and port
-    let host = 'localhost';
-    let port = FALKORDB_DEFAULT_PORT;
-
-    // Support both bolt:// and redis:// protocols
-    const uriMatch = uri.match(/^(?:bolt|redis):\/\/([^:]+):(\d+)/);
-    if (uriMatch) {
-      host = uriMatch[1];
-      port = parseInt(uriMatch[2], 10);
-    } else {
-      // Try simple host:port format
-      const simpleMatch = uri.match(/^([^:]+):(\d+)/);
-      if (simpleMatch) {
-        host = simpleMatch[1];
-        port = parseInt(simpleMatch[2], 10);
-      }
-    }
-
-    const startTime = Date.now();
-
-    // First, check the actual FalkorDB container status to get the correct port
-    const falkorStatus = await checkFalkorDBStatus(port);
-
-    // If container exists but user specified wrong port, try to detect the actual port
-    if (!falkorStatus.containerRunning) {
-      // Check if container is running on default port
-      const defaultStatus = await checkFalkorDBStatus(FALKORDB_DEFAULT_PORT);
-      if (defaultStatus.containerRunning && defaultStatus.healthy) {
-        return {
-          success: false,
-          message: `FalkorDB is running on port ${FALKORDB_DEFAULT_PORT}, but you specified port ${port}. Please update the URI to bolt://localhost:${FALKORDB_DEFAULT_PORT}`,
-        };
-      }
-
-      return {
-        success: false,
-        message: `FalkorDB container is not running. Please start FalkorDB first using Docker.`,
-      };
-    }
-
-    // Try to ping FalkorDB using redis-cli in Docker container
-    try {
-      const { stdout } = await execAsync(
-        `docker exec ${FALKORDB_CONTAINER_NAME} redis-cli PING`,
-        { timeout: 10000 }
-      );
-
-      if (stdout.trim().toUpperCase() === 'PONG') {
-        const latencyMs = Date.now() - startTime;
-        return {
-          success: true,
-          message: `Connected to FalkorDB at ${host}:${port}`,
-          details: { latencyMs },
-        };
-      }
-    } catch {
-      // redis-cli failed, try port check as fallback
-    }
-
-    // Fallback: check if the port is open using nc or direct connection
-    try {
-      // Check if we can connect to the mapped port from the host
-      await execAsync(`nc -z -w 5 ${host} ${port}`, { timeout: 10000 });
-      const latencyMs = Date.now() - startTime;
-      return {
-        success: true,
-        message: `FalkorDB port ${port} is reachable at ${host}`,
-        details: { latencyMs },
-      };
-    } catch {
-      // Port check failed, but container is running - might be a different port mapping
-      if (falkorStatus.containerRunning) {
-        return {
-          success: false,
-          message: `FalkorDB container is running but port ${port} is not reachable. The container may be mapped to a different port.`,
-        };
-      }
-
-      return {
-        success: false,
-        message: `Cannot connect to FalkorDB at ${host}:${port}. Make sure FalkorDB is running.`,
-      };
-    }
-  } catch (error) {
-    return {
-      success: false,
-      message: error instanceof Error ? error.message : 'Unknown error occurred',
-    };
-  }
-}
-
-/**
- * Validate OpenAI API key by attempting to list models
- * @param apiKey - OpenAI API key
- */
-export async function validateOpenAIApiKey(
-  apiKey: string
-): Promise<GraphitiValidationResult> {
-  if (!apiKey || !apiKey.trim()) {
-    return {
-      success: false,
-      message: 'API key is required',
-    };
-  }
-
-  // Basic format validation
-  const trimmedKey = apiKey.trim();
-  if (!trimmedKey.startsWith('sk-') && !trimmedKey.startsWith('sess-')) {
-    return {
-      success: false,
-      message: 'Invalid API key format. OpenAI API keys should start with "sk-"',
-    };
-  }
-
-  try {
-    const startTime = Date.now();
-
-    // Use native https module to avoid additional dependencies
-    const result = await new Promise<GraphitiValidationResult>((resolve) => {
-      const https = require('https');
-
-      const options = {
-        hostname: 'api.openai.com',
-        port: 443,
-        path: '/v1/models',
-        method: 'GET',
-        headers: {
-          Authorization: `Bearer ${trimmedKey}`,
-          'Content-Type': 'application/json',
-        },
-        timeout: 15000,
-      };
-
-      const req = https.request(options, (res: { statusCode: number; on: (event: string, callback: (chunk: Buffer) => void) => void }) => {
-        let data = '';
-
-        res.on('data', (chunk: Buffer) => {
-          data += chunk;
-        });
-
-        res.on('end', () => {
-          const latencyMs = Date.now() - startTime;
-
-          if (res.statusCode === 200) {
-            resolve({
-              success: true,
-              message: 'OpenAI API key is valid',
-              details: {
-                provider: 'openai',
-                latencyMs,
-              },
-            });
-          } else if (res.statusCode === 401) {
-            resolve({
-              success: false,
-              message: 'Invalid API key. Please check your OpenAI API key.',
-            });
-          } else if (res.statusCode === 429) {
-            // Rate limited but key is valid
-            resolve({
-              success: true,
-              message: 'OpenAI API key is valid (rate limited, please wait)',
-              details: {
-                provider: 'openai',
-                latencyMs,
-              },
-            });
-          } else {
-            try {
-              const errorData = JSON.parse(data);
-              resolve({
-                success: false,
-                message: errorData.error?.message || `API error: ${res.statusCode}`,
-              });
-            } catch {
-              resolve({
-                success: false,
-                message: `API error: ${res.statusCode}`,
-              });
-            }
-          }
-        });
-      });
-
-      req.on('error', (error: Error) => {
-        resolve({
-          success: false,
-          message: `Connection error: ${error.message}`,
-        });
-      });
-
-      req.on('timeout', () => {
-        req.destroy();
-        resolve({
-          success: false,
-          message: 'Connection timeout. Please check your network connection.',
-        });
-      });
-
-      req.end();
-    });
-
-    return result;
-  } catch (error) {
-    return {
-      success: false,
-      message: error instanceof Error ? error.message : 'Unknown error occurred',
-    };
-  }
-}
-
-/**
- * Test the full Graphiti connection (FalkorDB + OpenAI)
- * @param falkorDbUri - FalkorDB URI
- * @param openAiApiKey - OpenAI API key
- */
-export async function testGraphitiConnection(
-  falkorDbUri: string,
-  openAiApiKey: string
-): Promise<{
-  falkordb: GraphitiValidationResult;
-  openai: GraphitiValidationResult;
-  ready: boolean;
-}> {
-  const [falkordb, openai] = await Promise.all([
-    validateFalkorDBConnection(falkorDbUri),
-    validateOpenAIApiKey(openAiApiKey),
-  ]);
-
-  return {
-    falkordb,
-    openai,
-    ready: falkordb.success && openai.success,
-  };
-}

auto-claude-ui/src/main/falkordb-service.ts

@@ -1,365 +0,0 @@
-/**
- * FalkorDB Service
- *
- * Queries the FalkorDB graph database for memories stored by Graphiti.
- * Uses ioredis to communicate with FalkorDB via Redis protocol.
- */
-
-import Redis from 'ioredis';
-import type { MemoryEpisode } from '../shared/types';
-
-interface FalkorDBConfig {
-  host: string;
-  port: number;
-  password?: string;
-}
-
-interface EpisodicNode {
-  uuid: string;
-  name: string;
-  created_at: string;
-  content?: string;
-  source_description?: string;
-}
-
-interface EntityNode {
-  uuid: string;
-  name: string;
-  summary?: string;
-}
-
-/**
- * Parse FalkorDB GRAPH.QUERY results into structured data
- */
-function parseGraphResult(result: unknown[]): Record<string, unknown>[] {
-  if (!Array.isArray(result) || result.length < 2) {
-    return [];
-  }
-
-  // Result format: [headers, [row1, row2, ...], stats]
-  const headers = result[0] as string[];
-  const rows = result[1] as unknown[][];
-
-  if (!Array.isArray(headers) || !Array.isArray(rows)) {
-    return [];
-  }
-
-  return rows.map(row => {
-    const obj: Record<string, unknown> = {};
-    headers.forEach((header, idx) => {
-      obj[header] = row[idx];
-    });
-    return obj;
-  });
-}
-
-/**
- * FalkorDB Service for querying graph memories
- */
-export class FalkorDBService {
-  private config: FalkorDBConfig;
-  private redis: Redis | null = null;
-
-  constructor(config: FalkorDBConfig) {
-    this.config = config;
-  }
-
-  /**
-   * Get a Redis connection (lazy initialization)
-   */
-  private async getConnection(): Promise<Redis> {
-    if (this.redis) {
-      return this.redis;
-    }
-
-    this.redis = new Redis({
-      host: this.config.host,
-      port: this.config.port,
-      password: this.config.password,
-      lazyConnect: true,
-      connectTimeout: 5000,
-      maxRetriesPerRequest: 1,
-    });
-
-    await this.redis.connect();
-    return this.redis;
-  }
-
-  /**
-   * Close the Redis connection
-   */
-  async close(): Promise<void> {
-    if (this.redis) {
-      await this.redis.quit();
-      this.redis = null;
-    }
-  }
-
-  /**
-   * List all available graphs in the database
-   */
-  async listGraphs(): Promise<string[]> {
-    try {
-      const redis = await this.getConnection();
-      const result = await redis.call('GRAPH.LIST') as string[];
-      return result || [];
-    } catch (error) {
-      console.error('Failed to list graphs:', error);
-      return [];
-    }
-  }
-
-  /**
-   * Query episodic memories from a specific graph
-   */
-  async getEpisodicMemories(graphName: string, limit: number = 20): Promise<MemoryEpisode[]> {
-    try {
-      const redis = await this.getConnection();
-
-      // Query episodic nodes with their details
-      const query = `
-        MATCH (e:Episodic)
-        RETURN e.uuid as uuid, e.name as name, e.created_at as created_at,
-               e.content as content, e.source_description as description
-        ORDER BY e.created_at DESC
-        LIMIT ${limit}
-      `;
-
-      const result = await redis.call('GRAPH.QUERY', graphName, query) as unknown[];
-      const episodes = parseGraphResult(result) as unknown as EpisodicNode[];
-
-      return episodes.map(ep => ({
-        id: ep.uuid || ep.name,
-        type: this.inferEpisodeType(ep.name, ep.content),
-        timestamp: ep.created_at || new Date().toISOString(),
-        content: ep.content || ep.source_description || ep.name,
-        session_number: this.extractSessionNumber(ep.name),
-      }));
-    } catch (error) {
-      console.error(`Failed to get episodic memories from ${graphName}:`, error);
-      return [];
-    }
-  }
-
-  /**
-   * Query entity memories (patterns, gotchas, etc.) from a graph
-   */
-  async getEntityMemories(graphName: string, limit: number = 20): Promise<MemoryEpisode[]> {
-    try {
-      const redis = await this.getConnection();
-
-      // Query entity nodes
-      const query = `
-        MATCH (e:Entity)
-        RETURN e.uuid as uuid, e.name as name, e.summary as summary, e.created_at as created_at
-        ORDER BY e.created_at DESC
-        LIMIT ${limit}
-      `;
-
-      const result = await redis.call('GRAPH.QUERY', graphName, query) as unknown[];
-      const entities = parseGraphResult(result) as unknown as EntityNode[];
-
-      return entities
-        .filter(ent => ent.summary) // Only include entities with summaries
-        .map(ent => ({
-          id: ent.uuid || ent.name,
-          type: this.inferEntityType(ent.name),
-          timestamp: new Date().toISOString(),
-          content: ent.summary || ent.name,
-        }));
-    } catch (error) {
-      console.error(`Failed to get entity memories from ${graphName}:`, error);
-      return [];
-    }
-  }
-
-  /**
-   * Get all memories from all spec-related graphs
-   */
-  async getAllMemories(limit: number = 20): Promise<MemoryEpisode[]> {
-    const graphs = await this.listGraphs();
-    const memories: MemoryEpisode[] = [];
-
-    // Filter to spec-related graphs (exclude auto_build_memory and project_ prefixed)
-    const specGraphs = graphs.filter(g =>
-      !g.startsWith('project_') &&
-      g !== 'auto_build_memory' &&
-      g !== 'default_db'
-    );
-
-    for (const graph of specGraphs) {
-      const episodic = await this.getEpisodicMemories(graph, Math.ceil(limit / specGraphs.length));
-      memories.push(...episodic.map(m => ({ ...m, id: `${graph}:${m.id}` })));
-    }
-
-    // Sort by timestamp descending
-    memories.sort((a, b) => new Date(b.timestamp).getTime() - new Date(a.timestamp).getTime());
-
-    return memories.slice(0, limit);
-  }
-
-  /**
-   * Search memories across all graphs
-   */
-  async searchMemories(query: string, limit: number = 20): Promise<MemoryEpisode[]> {
-    const graphs = await this.listGraphs();
-    const results: MemoryEpisode[] = [];
-    const queryLower = query.toLowerCase();
-
-    // Filter to spec-related graphs
-    const specGraphs = graphs.filter(g =>
-      !g.startsWith('project_') &&
-      g !== 'auto_build_memory' &&
-      g !== 'default_db'
-    );
-
-    for (const graph of specGraphs) {
-      try {
-        const redis = await this.getConnection();
-
-        // Search in episodic nodes
-        const episodicQuery = `
-          MATCH (e:Episodic)
-          WHERE toLower(e.name) CONTAINS '${queryLower}' OR toLower(e.content) CONTAINS '${queryLower}'
-          RETURN e.uuid as uuid, e.name as name, e.created_at as created_at,
-                 e.content as content, e.source_description as description
-          LIMIT ${Math.ceil(limit / specGraphs.length)}
-        `;
-
-        const episodicResult = await redis.call('GRAPH.QUERY', graph, episodicQuery) as unknown[];
-        const episodes = parseGraphResult(episodicResult) as unknown as EpisodicNode[];
-
-        results.push(...episodes.map(ep => ({
-          id: `${graph}:${ep.uuid || ep.name}`,
-          type: this.inferEpisodeType(ep.name, ep.content),
-          timestamp: ep.created_at || new Date().toISOString(),
-          content: ep.content || ep.source_description || ep.name,
-          session_number: this.extractSessionNumber(ep.name),
-          score: 1.0,
-        })));
-
-        // Search in entity nodes
-        const entityQuery = `
-          MATCH (e:Entity)
-          WHERE toLower(e.name) CONTAINS '${queryLower}' OR toLower(e.summary) CONTAINS '${queryLower}'
-          RETURN e.uuid as uuid, e.name as name, e.summary as summary
-          LIMIT ${Math.ceil(limit / specGraphs.length)}
-        `;
-
-        const entityResult = await redis.call('GRAPH.QUERY', graph, entityQuery) as unknown[];
-        const entities = parseGraphResult(entityResult) as unknown as EntityNode[];
-
-        results.push(...entities
-          .filter(ent => ent.summary)
-          .map(ent => ({
-            id: `${graph}:${ent.uuid || ent.name}`,
-            type: this.inferEntityType(ent.name),
-            timestamp: new Date().toISOString(),
-            content: ent.summary || ent.name,
-            score: 1.0,
-          })));
-      } catch (error) {
-        console.error(`Failed to search memories in ${graph}:`, error);
-      }
-    }
-
-    return results.slice(0, limit);
-  }
-
-  /**
-   * Test connection to FalkorDB
-   */
-  async testConnection(): Promise<{ success: boolean; message: string }> {
-    try {
-      const redis = await this.getConnection();
-      await redis.ping();
-      const graphs = await this.listGraphs();
-      return {
-        success: true,
-        message: `Connected to FalkorDB with ${graphs.length} graphs`,
-      };
-    } catch (error) {
-      return {
-        success: false,
-        message: error instanceof Error ? error.message : 'Connection failed',
-      };
-    }
-  }
-
-  /**
-   * Infer the episode type from its name
-   */
-  private inferEpisodeType(name: string, content?: string): MemoryEpisode['type'] {
-    const nameLower = (name || '').toLowerCase();
-    const contentLower = (content || '').toLowerCase();
-
-    if (nameLower.includes('session_') || contentLower.includes('"type": "session_insight"')) {
-      return 'session_insight';
-    }
-    if (nameLower.includes('pattern') || contentLower.includes('"type": "pattern"')) {
-      return 'pattern';
-    }
-    if (nameLower.includes('gotcha') || contentLower.includes('"type": "gotcha"')) {
-      return 'gotcha';
-    }
-    if (nameLower.includes('codebase') || contentLower.includes('"type": "codebase_discovery"')) {
-      return 'codebase_discovery';
-    }
-    return 'session_insight';
-  }
-
-  /**
-   * Infer the entity type from its name
-   */
-  private inferEntityType(name: string): MemoryEpisode['type'] {
-    const nameLower = (name || '').toLowerCase();
-
-    if (nameLower.includes('pattern')) {
-      return 'pattern';
-    }
-    if (nameLower.includes('gotcha')) {
-      return 'gotcha';
-    }
-    if (nameLower.includes('file_insight') || nameLower.includes('codebase')) {
-      return 'codebase_discovery';
-    }
-    return 'session_insight';
-  }
-
-  /**
-   * Extract session number from episode name
-   */
-  private extractSessionNumber(name: string): number | undefined {
-    const match = name.match(/session_(\d+)/i);
-    return match ? parseInt(match[1], 10) : undefined;
-  }
-}
-
-// Singleton instance for reuse
-let serviceInstance: FalkorDBService | null = null;
-
-/**
- * Get or create a FalkorDB service instance
- */
-export function getFalkorDBService(config: FalkorDBConfig): FalkorDBService {
-  if (!serviceInstance ||
-      serviceInstance['config'].host !== config.host ||
-      serviceInstance['config'].port !== config.port) {
-    // Close existing connection if config changed
-    if (serviceInstance) {
-      serviceInstance.close().catch(() => {});
-    }
-    serviceInstance = new FalkorDBService(config);
-  }
-  return serviceInstance;
-}
-
-/**
- * Close the singleton service instance
- */
-export async function closeFalkorDBService(): Promise<void> {
-  if (serviceInstance) {
-    await serviceInstance.close();
-    serviceInstance = null;
-  }
-}

auto-claude-ui/src/main/index.ts

@@ -141,16 +141,9 @@ app.whenReady().then(() => {
 
     // Log debug mode status
     const isDebugMode = process.env.DEBUG === 'true';
-    const isAutoClaudeDebug = process.env.AUTO_CLAUDE_DEBUG === 'true';
-    if (isDebugMode || isAutoClaudeDebug) {
+    if (isDebugMode) {
       console.warn('[main] ========================================');
-      console.warn('[main] DEBUG MODE ENABLED');
-      if (isDebugMode) {
-        console.warn('[main] - DEBUG=true (Ideation/Roadmap debug logging)');
-      }
-      if (isAutoClaudeDebug) {
-        console.warn('[main] - AUTO_CLAUDE_DEBUG=true (Core features debug logging)');
-      }
+      console.warn('[main] DEBUG MODE ENABLED (DEBUG=true)');
       console.warn('[main] ========================================');
     }
 

auto-claude-ui/src/main/insights/config.ts

@@ -2,13 +2,15 @@ import path from 'path';
 import { existsSync, readFileSync } from 'fs';
 import { app } from 'electron';
 import { getProfileEnv } from '../rate-limit-detector';
+import { findPythonCommand } from '../python-detector';
 
 /**
  * Configuration manager for insights service
  * Handles path detection and environment variable loading
  */
 export class InsightsConfig {
-  private pythonPath: string = 'python3';
+  // Auto-detect Python command on initialization
+  private pythonPath: string = findPythonCommand() || 'python';
   private autoBuildSourcePath: string = '';
 
   /**

auto-claude-ui/src/main/insights/insights-executor.ts

@@ -1,6 +1,7 @@
 import { spawn, ChildProcess } from 'child_process';
-import { existsSync } from 'fs';
+import { existsSync, writeFileSync, unlinkSync } from 'fs';
 import path from 'path';
+import os from 'os';
 import { EventEmitter } from 'events';
 import type {
   InsightsChatMessage,
@@ -86,12 +87,27 @@ export class InsightsExecutor extends EventEmitter {
     // Get process environment
     const processEnv = this.config.getProcessEnv();
 
+    // Write conversation history to temp file to avoid Windows command-line length limit
+    const historyFile = path.join(
+      os.tmpdir(),
+      `insights-history-${projectId}-${Date.now()}.json`
+    );
+
+    let historyFileCreated = false;
+    try {
+      writeFileSync(historyFile, JSON.stringify(conversationHistory), 'utf-8');
+      historyFileCreated = true;
+    } catch (err) {
+      console.error('[Insights] Failed to write history file:', err);
+      throw new Error('Failed to write conversation history to temp file');
+    }
+
     // Build command arguments
     const args = [
       runnerPath,
       '--project-dir', projectPath,
       '--message', message,
-      '--history', JSON.stringify(conversationHistory)
+      '--history-file', historyFile
     ];
 
     // Add model config if provided
@@ -151,6 +167,15 @@ export class InsightsExecutor extends EventEmitter {
       proc.on('close', (code) => {
         this.activeSessions.delete(projectId);
 
+        // Cleanup temp file
+        if (historyFileCreated && existsSync(historyFile)) {
+          try {
+            unlinkSync(historyFile);
+          } catch (cleanupErr) {
+            console.error('[Insights] Failed to cleanup history file:', cleanupErr);
+          }
+        }
+
         // Check for rate limit if process failed
         if (code !== 0) {
           this.handleRateLimit(projectId, allInsightsOutput);
@@ -184,6 +209,16 @@ export class InsightsExecutor extends EventEmitter {
 
       proc.on('error', (err) => {
         this.activeSessions.delete(projectId);
+
+        // Cleanup temp file
+        if (historyFileCreated && existsSync(historyFile)) {
+          try {
+            unlinkSync(historyFile);
+          } catch (cleanupErr) {
+            console.error('[Insights] Failed to cleanup history file:', cleanupErr);
+          }
+        }
+
         this.emit('error', projectId, err.message);
         reject(err);
       });

auto-claude-ui/src/main/integrations/index.ts

@@ -0,0 +1,18 @@
+/**
+ * Integration module for external roadmap/feedback services
+ *
+ * Currently provides architecture for future integrations with:
+ * - Canny.io (feedback management)
+ * - GitHub Issues
+ *
+ * To add a new integration:
+ * 1. Implement the IntegrationAdapter interface
+ * 2. Add status mapping constants
+ * 3. Register the adapter in this module
+ */
+
+export * from './types';
+
+// Future: Export concrete adapter implementations
+// export { CannyAdapter } from './canny-adapter';
+// export { GitHubIssuesAdapter } from './github-issues-adapter';

auto-claude-ui/src/main/integrations/types.ts

@@ -0,0 +1,125 @@
+/**
+ * Integration provider types for external roadmap services (Canny, GitHub Issues, etc.)
+ *
+ * This architecture allows bidirectional sync with external feedback/roadmap systems:
+ * - Import: Fetch feature requests from external services
+ * - Export: Push status updates back when features progress
+ */
+
+import type { RoadmapFeatureStatus } from '../../shared/types';
+
+/**
+ * Represents an item from an external feedback/roadmap system
+ */
+export interface FeedbackItem {
+  externalId: string;
+  title: string;
+  description: string;
+  votes: number;
+  status: string;  // Provider-specific status
+  url: string;
+  createdAt: Date;
+  updatedAt?: Date;
+  author?: string;
+  tags?: string[];
+}
+
+/**
+ * Connection status for a provider
+ */
+export interface ProviderConnection {
+  id: string;
+  name: string;
+  connected: boolean;
+  lastSync?: Date;
+  error?: string;
+}
+
+/**
+ * Configuration for a provider
+ */
+export interface ProviderConfig {
+  enabled: boolean;
+  apiKey?: string;
+  boardId?: string;
+  autoSync?: boolean;
+  syncIntervalMinutes?: number;
+}
+
+/**
+ * Abstract interface for integration adapters
+ *
+ * Implement this interface to add support for new external services.
+ * Each adapter handles mapping between internal and external status systems.
+ */
+export interface IntegrationAdapter {
+  /** Unique identifier for this provider */
+  readonly providerId: string;
+
+  /** Display name for the provider */
+  readonly providerName: string;
+
+  /**
+   * Test the connection to the external service
+   */
+  testConnection(): Promise<{ success: boolean; error?: string }>;
+
+  /**
+   * Fetch all items from the external service
+   */
+  fetchItems(): Promise<FeedbackItem[]>;
+
+  /**
+   * Update the status of an item in the external service
+   */
+  updateStatus(externalId: string, status: string): Promise<void>;
+
+  /**
+   * Map internal roadmap status to provider-specific status
+   */
+  mapStatusToProvider(internalStatus: RoadmapFeatureStatus): string;
+
+  /**
+   * Map provider-specific status to internal roadmap status
+   */
+  mapStatusFromProvider(externalStatus: string): RoadmapFeatureStatus;
+}
+
+/**
+ * Canny-specific status mapping
+ * Reference: https://developers.canny.io/api-reference
+ */
+export const CANNY_STATUS_MAP = {
+  toProvider: {
+    under_review: 'under review',
+    planned: 'planned',
+    in_progress: 'in progress',
+    done: 'complete'
+  } as Record<RoadmapFeatureStatus, string>,
+
+  fromProvider: {
+    'open': 'under_review',
+    'under review': 'under_review',
+    'planned': 'planned',
+    'in progress': 'in_progress',
+    'complete': 'done',
+    'closed': 'done'
+  } as Record<string, RoadmapFeatureStatus>
+};
+
+/**
+ * GitHub Issues status mapping
+ */
+export const GITHUB_ISSUES_STATUS_MAP = {
+  toProvider: {
+    under_review: 'open',
+    planned: 'open',
+    in_progress: 'open',
+    done: 'closed'
+  } as Record<RoadmapFeatureStatus, string>,
+
+  fromProvider: {
+    'open': 'under_review',
+    'closed': 'done'
+  } as Record<string, RoadmapFeatureStatus>
+};

auto-claude-ui/src/main/ipc-handlers.ts.backup

@@ -1256,7 +1256,7 @@ export function setupIpcHandlers(
           try {
             // Set status to in_progress for the restart
             newStatus = 'in_progress';
-            
+
             // Update plan status for restart
             if (plan) {
               plan.status = 'in_progress';
@@ -1277,7 +1277,7 @@ export function setupIpcHandlers(
             const specsBaseDir = getSpecsDir(project.autoBuildPath);
             const specDirForWatcher = path.join(project.path, specsBaseDir, task.specId);
             fileWatcher.watch(taskId, specDirForWatcher);
-            
+
             agentManager.startTaskExecution(
               taskId,
               project.path,
@@ -1312,7 +1312,7 @@ export function setupIpcHandlers(
             taskId,
             recovered: true,
             newStatus,
-            message: autoRestarted 
+            message: autoRestarted
               ? 'Task recovered and restarted successfully'
               : `Task recovered successfully and moved to ${newStatus}`,
             autoRestarted
@@ -2418,12 +2418,12 @@ export function setupIpcHandlers(
           });
         }
 
-        return { 
-          success: true, 
-          data: { 
+        return {
+          success: true,
+          data: {
             terminalId,
             message: `A terminal has been opened to authenticate "${profile.name}". Complete the OAuth flow in your browser, then copy the token shown in the terminal.`
-          } 
+          }
         };
       } catch (error) {
         console.error('[IPC] Failed to initialize Claude profile:', error);

auto-claude-ui/src/main/ipc-handlers/autobuild-source-handlers.ts

@@ -5,7 +5,8 @@ import type { IPCResult } from '../../shared/types';
 import path from 'path';
 import { existsSync, readFileSync, writeFileSync } from 'fs';
 import type { AutoBuildSourceUpdateProgress, SourceEnvConfig, SourceEnvCheckResult } from '../../shared/types';
-import { checkForUpdates as checkSourceUpdates, downloadAndApplyUpdate, getBundledVersion, getEffectiveSourcePath } from '../auto-claude-updater';
+import { checkForUpdates as checkSourceUpdates, downloadAndApplyUpdate, getBundledVersion, getEffectiveVersion, getEffectiveSourcePath } from '../auto-claude-updater';
+import { debugLog } from '../../shared/utils/debug-logger';
 
 
 /**
@@ -21,10 +22,16 @@ export function registerAutobuildSourceHandlers(
   ipcMain.handle(
     IPC_CHANNELS.AUTOBUILD_SOURCE_CHECK,
     async (): Promise<IPCResult<{ updateAvailable: boolean; currentVersion: string; latestVersion?: string; releaseNotes?: string; releaseUrl?: string; error?: string }>> => {
+      console.log('[autobuild-source] Check for updates called');
+      debugLog('[IPC] AUTOBUILD_SOURCE_CHECK called');
       try {
         const result = await checkSourceUpdates();
+        console.log('[autobuild-source] Check result:', JSON.stringify(result, null, 2));
+        debugLog('[IPC] AUTOBUILD_SOURCE_CHECK result:', result);
         return { success: true, data: result };
       } catch (error) {
+        console.error('[autobuild-source] Check error:', error);
+        debugLog('[IPC] AUTOBUILD_SOURCE_CHECK error:', error);
         return {
           success: false,
           error: error instanceof Error ? error.message : 'Failed to check for updates'
@@ -36,25 +43,33 @@ export function registerAutobuildSourceHandlers(
   ipcMain.on(
     IPC_CHANNELS.AUTOBUILD_SOURCE_DOWNLOAD,
     () => {
+      debugLog('[IPC] Autobuild source download requested');
       const mainWindow = getMainWindow();
-      if (!mainWindow) return;
+      if (!mainWindow) {
+        debugLog('[IPC] No main window available, aborting update');
+        return;
+      }
 
       // Start download in background
       downloadAndApplyUpdate((progress) => {
+        debugLog('[IPC] Update progress:', progress.stage, progress.message);
         mainWindow.webContents.send(
           IPC_CHANNELS.AUTOBUILD_SOURCE_PROGRESS,
           progress
         );
       }).then((result) => {
         if (result.success) {
+          debugLog('[IPC] Update completed successfully, version:', result.version);
           mainWindow.webContents.send(
             IPC_CHANNELS.AUTOBUILD_SOURCE_PROGRESS,
             {
               stage: 'complete',
-              message: `Updated to version ${result.version}`
+              message: `Updated to version ${result.version}`,
+              newVersion: result.version // Include new version for UI refresh
             } as AutoBuildSourceUpdateProgress
           );
         } else {
+          debugLog('[IPC] Update failed:', result.error);
           mainWindow.webContents.send(
             IPC_CHANNELS.AUTOBUILD_SOURCE_PROGRESS,
             {
@@ -64,6 +79,7 @@ export function registerAutobuildSourceHandlers(
           );
         }
       }).catch((error) => {
+        debugLog('[IPC] Update error:', error instanceof Error ? error.message : error);
         mainWindow.webContents.send(
           IPC_CHANNELS.AUTOBUILD_SOURCE_PROGRESS,
           {
@@ -88,7 +104,9 @@ export function registerAutobuildSourceHandlers(
     IPC_CHANNELS.AUTOBUILD_SOURCE_VERSION,
     async (): Promise<IPCResult<string>> => {
       try {
-        const version = getBundledVersion();
+        // Use effective version which accounts for source updates
+        const version = getEffectiveVersion();
+        debugLog('[IPC] Returning effective version:', version);
         return { success: true, data: version };
       } catch (error) {
         return {

auto-claude-ui/src/main/ipc-handlers/changelog-handlers.ts

@@ -1,7 +1,7 @@
 import { ipcMain } from 'electron';
 import type { BrowserWindow } from 'electron';
 import path from 'path';
-import { existsSync, mkdirSync, writeFileSync } from 'fs';
+import { existsSync, mkdirSync, writeFileSync, readFileSync } from 'fs';
 import { IPC_CHANNELS, getSpecsDir } from '../../shared/constants';
 import type {
   IPCResult,
@@ -375,6 +375,46 @@ export function registerChangelogHandlers(
     }
   );
 
+  ipcMain.handle(
+    IPC_CHANNELS.CHANGELOG_READ_LOCAL_IMAGE,
+    async (_, projectPath: string, relativePath: string): Promise<IPCResult<string>> => {
+      try {
+        // Construct full path from project path and relative path
+        const fullPath = path.join(projectPath, relativePath);
+
+        // Verify the file exists
+        if (!existsSync(fullPath)) {
+          return { success: false, error: `Image not found: ${relativePath}` };
+        }
+
+        // Read the file and convert to base64
+        const buffer = readFileSync(fullPath);
+        const base64 = buffer.toString('base64');
+
+        // Determine MIME type from extension
+        const ext = path.extname(relativePath).toLowerCase();
+        const mimeTypes: Record<string, string> = {
+          '.png': 'image/png',
+          '.jpg': 'image/jpeg',
+          '.jpeg': 'image/jpeg',
+          '.gif': 'image/gif',
+          '.webp': 'image/webp',
+          '.svg': 'image/svg+xml'
+        };
+        const mimeType = mimeTypes[ext] || 'image/png';
+
+        // Return as data URL
+        const dataUrl = `data:${mimeType};base64,${base64}`;
+        return { success: true, data: dataUrl };
+      } catch (error) {
+        return {
+          success: false,
+          error: error instanceof Error ? error.message : 'Failed to read image'
+        };
+      }
+    }
+  );
+
   // ============================================
   // Changelog Agent Events → Renderer
 }

auto-claude-ui/src/main/ipc-handlers/context/memory-data-handlers.ts

@@ -9,11 +9,11 @@ import type {
   ContextSearchResult
 } from '../../../shared/types';
 import { projectStore } from '../../project-store';
-import { getFalkorDBService } from '../../falkordb-service';
+import { getMemoryService, isKuzuAvailable } from '../../memory-service';
 import {
   loadProjectEnvVars,
   isGraphitiEnabled,
-  getGraphitiConnectionDetails
+  getGraphitiDatabaseDetails
 } from './utils';
 
 /**
@@ -169,20 +169,20 @@ export function registerMemoryDataHandlers(
       const projectEnvVars = loadProjectEnvVars(project.path, project.autoBuildPath);
       const graphitiEnabled = isGraphitiEnabled(projectEnvVars);
 
-      // Try FalkorDB first if available
-      if (graphitiEnabled) {
+      // Try LadybugDB first if available
+      if (graphitiEnabled && isKuzuAvailable()) {
         try {
-          const connDetails = getGraphitiConnectionDetails(projectEnvVars);
-          const falkorService = getFalkorDBService({
-            host: connDetails.host,
-            port: connDetails.port,
+          const dbDetails = getGraphitiDatabaseDetails(projectEnvVars);
+          const memoryService = getMemoryService({
+            dbPath: dbDetails.dbPath,
+            database: dbDetails.database,
           });
-          const falkorMemories = await falkorService.getAllMemories(limit);
-          if (falkorMemories.length > 0) {
-            return { success: true, data: falkorMemories };
+          const graphMemories = await memoryService.getEpisodicMemories(limit);
+          if (graphMemories.length > 0) {
+            return { success: true, data: graphMemories };
           }
         } catch (error) {
-          console.warn('Failed to get memories from FalkorDB, falling back to file-based:', error);
+          console.warn('Failed to get memories from LadybugDB, falling back to file-based:', error);
         }
       }
 
@@ -207,19 +207,19 @@ export function registerMemoryDataHandlers(
       const projectEnvVars = loadProjectEnvVars(project.path, project.autoBuildPath);
       const graphitiEnabled = isGraphitiEnabled(projectEnvVars);
 
-      // Try FalkorDB search if available
-      if (graphitiEnabled) {
+      // Try LadybugDB search if available
+      if (graphitiEnabled && isKuzuAvailable()) {
         try {
-          const connDetails = getGraphitiConnectionDetails(projectEnvVars);
-          const falkorService = getFalkorDBService({
-            host: connDetails.host,
-            port: connDetails.port,
+          const dbDetails = getGraphitiDatabaseDetails(projectEnvVars);
+          const memoryService = getMemoryService({
+            dbPath: dbDetails.dbPath,
+            database: dbDetails.database,
           });
-          const falkorResults = await falkorService.searchMemories(query, 20);
-          if (falkorResults.length > 0) {
+          const graphResults = await memoryService.searchMemories(query, 20);
+          if (graphResults.length > 0) {
             return {
               success: true,
-              data: falkorResults.map(r => ({
+              data: graphResults.map(r => ({
                 content: r.content,
                 score: r.score || 1.0,
                 type: r.type
@@ -227,7 +227,7 @@ export function registerMemoryDataHandlers(
             };
           }
         } catch (error) {
-          console.warn('Failed to search FalkorDB, falling back to file-based:', error);
+          console.warn('Failed to search LadybugDB, falling back to file-based:', error);
         }
       }
 

auto-claude-ui/src/main/ipc-handlers/context/memory-status-handlers.ts

@@ -10,7 +10,7 @@ import {
   loadGlobalSettings,
   isGraphitiEnabled,
   hasOpenAIKey,
-  getGraphitiConnectionDetails
+  getGraphitiDatabaseDetails
 } from './utils';
 
 /**
@@ -65,13 +65,12 @@ export function buildMemoryStatus(
 
   // If we have initialized state from specs, use it
   if (memoryState?.initialized) {
-    const connDetails = getGraphitiConnectionDetails(projectEnvVars);
+    const dbDetails = getGraphitiDatabaseDetails(projectEnvVars);
     return {
       enabled: true,
       available: true,
       database: memoryState.database || 'auto_claude_memory',
-      host: connDetails.host,
-      port: connDetails.port
+      dbPath: dbDetails.dbPath
     };
   }
 
@@ -95,13 +94,12 @@ export function buildMemoryStatus(
     };
   }
 
-  const connDetails = getGraphitiConnectionDetails(projectEnvVars);
+  const dbDetails = getGraphitiDatabaseDetails(projectEnvVars);
   return {
     enabled: true,
     available: true,
-    host: connDetails.host,
-    port: connDetails.port,
-    database: connDetails.database
+    dbPath: dbDetails.dbPath,
+    database: dbDetails.database
   };
 }
 

auto-claude-ui/src/main/ipc-handlers/context/project-context-handlers.ts

@@ -11,10 +11,11 @@ import type {
   MemoryEpisode
 } from '../../../shared/types';
 import { projectStore } from '../../project-store';
-import { getFalkorDBService } from '../../falkordb-service';
+import { getMemoryService, isKuzuAvailable } from '../../memory-service';
 import {
-  getAutoBuildSourcePath
+  getGraphitiDatabaseDetails
 } from './utils';
+import { getEffectiveSourcePath } from '../../updater/path-resolver';
 import {
   loadGraphitiStateFromSpecs,
   buildMemoryStatus
@@ -39,34 +40,34 @@ function loadProjectIndex(projectPath: string): ProjectIndex | null {
 }
 
 /**
- * Load recent memories with FalkorDB fallback
+ * Load recent memories from LadybugDB with file-based fallback
  */
 async function loadRecentMemories(
   projectPath: string,
   autoBuildPath: string | undefined,
   memoryStatusAvailable: boolean,
-  memoryHost?: string,
-  memoryPort?: number
+  dbPath?: string,
+  database?: string
 ): Promise<MemoryEpisode[]> {
   let recentMemories: MemoryEpisode[] = [];
 
-  // Try to load from FalkorDB first if Graphiti is available
-  if (memoryStatusAvailable && memoryHost && memoryPort) {
+  // Try to load from LadybugDB first if Graphiti is available and Kuzu is installed
+  if (memoryStatusAvailable && isKuzuAvailable() && dbPath && database) {
     try {
-      const falkorService = getFalkorDBService({
-        host: memoryHost,
-        port: memoryPort,
+      const memoryService = getMemoryService({
+        dbPath,
+        database,
       });
-      const falkorMemories = await falkorService.getAllMemories(20);
-      if (falkorMemories.length > 0) {
-        recentMemories = falkorMemories;
+      const graphMemories = await memoryService.getEpisodicMemories(20);
+      if (graphMemories.length > 0) {
+        recentMemories = graphMemories;
       }
     } catch (error) {
-      console.warn('Failed to load memories from FalkorDB, falling back to file-based:', error);
+      console.warn('Failed to load memories from LadybugDB, falling back to file-based:', error);
     }
   }
 
-  // Fall back to file-based memory if no FalkorDB memories found
+  // Fall back to file-based memory if no graph memories found
   if (recentMemories.length === 0) {
     const specsBaseDir = getSpecsDir(autoBuildPath);
     const specsDir = path.join(projectPath, specsBaseDir);
@@ -110,8 +111,8 @@ export function registerProjectContextHandlers(
           project.path,
           project.autoBuildPath,
           memoryStatus.available,
-          memoryStatus.host,
-          memoryStatus.port
+          memoryStatus.dbPath,
+          memoryStatus.database
         );
 
         return {
@@ -144,7 +145,7 @@ export function registerProjectContextHandlers(
 
       try {
         // Run the analyzer script to regenerate project_index.json
-        const autoBuildSource = getAutoBuildSourcePath();
+        const autoBuildSource = getEffectiveSourcePath();
 
         if (!autoBuildSource) {
           return {

auto-claude-ui/src/main/ipc-handlers/context/utils.ts

@@ -120,29 +120,21 @@ export function hasOpenAIKey(projectEnvVars: EnvironmentVars, globalSettings: Gl
 }
 
 /**
- * Get Graphiti connection details
+ * Get Graphiti database details (LadybugDB - embedded database)
  */
-export interface GraphitiConnectionDetails {
-  host: string;
-  port: number;
+export interface GraphitiDatabaseDetails {
+  dbPath: string;
   database: string;
 }
 
-export function getGraphitiConnectionDetails(projectEnvVars: EnvironmentVars): GraphitiConnectionDetails {
-  const host = projectEnvVars['GRAPHITI_FALKORDB_HOST'] ||
-               process.env.GRAPHITI_FALKORDB_HOST ||
-               'localhost';
-
-  const port = parseInt(
-    projectEnvVars['GRAPHITI_FALKORDB_PORT'] ||
-    process.env.GRAPHITI_FALKORDB_PORT ||
-    '6380',
-    10
-  );
+export function getGraphitiDatabaseDetails(projectEnvVars: EnvironmentVars): GraphitiDatabaseDetails {
+  const dbPath = projectEnvVars['GRAPHITI_DB_PATH'] ||
+                 process.env.GRAPHITI_DB_PATH ||
+                 require('path').join(require('os').homedir(), '.auto-claude', 'memories');
 
   const database = projectEnvVars['GRAPHITI_DATABASE'] ||
                    process.env.GRAPHITI_DATABASE ||
                    'auto_claude_memory';
 
-  return { host, port, database };
+  return { dbPath, database };
 }

auto-claude-ui/src/main/ipc-handlers/docker-handlers.ts

@@ -1,156 +1,20 @@
 /**
- * Docker & FalkorDB IPC Handlers
+ * Docker & Infrastructure IPC Handlers
  *
- * Provides automatic infrastructure detection for non-technical users.
- * When Graphiti is enabled, the UI can check Docker/FalkorDB status
- * and offer one-click solutions instead of manual terminal commands.
+ * DEPRECATED: This file is kept for backward compatibility.
+ * Memory infrastructure has moved to LadybugDB (no Docker required).
+ * See memory-handlers.ts for the new implementation.
+ *
+ * This file now re-exports from memory-handlers.ts
  */
 
-import { ipcMain } from 'electron';
-import { IPC_CHANNELS } from '../../shared/constants';
-import type { IPCResult, InfrastructureStatus } from '../../shared/types';
-import {
-  getInfrastructureStatus,
-  startFalkorDB,
-  stopFalkorDB,
-  openDockerDesktop,
-  getDockerDownloadUrl,
-  validateFalkorDBConnection,
-  validateOpenAIApiKey,
-  testGraphitiConnection,
-  type GraphitiValidationResult,
-} from '../docker-service';
+import { registerMemoryHandlers } from './memory-handlers';
 
 /**
  * Register all Docker-related IPC handlers
+ * @deprecated Use registerMemoryHandlers() instead
  */
 export function registerDockerHandlers(): void {
-  // Get infrastructure status (Docker + FalkorDB)
-  ipcMain.handle(
-    IPC_CHANNELS.DOCKER_STATUS,
-    async (_, port?: number): Promise<IPCResult<InfrastructureStatus>> => {
-      try {
-        const status = await getInfrastructureStatus(port);
-        return { success: true, data: status };
-      } catch (error) {
-        return {
-          success: false,
-          error: error instanceof Error ? error.message : 'Failed to check Docker status',
-        };
-      }
-    }
-  );
-
-  // Start FalkorDB container
-  ipcMain.handle(
-    IPC_CHANNELS.DOCKER_START_FALKORDB,
-    async (_, port?: number): Promise<IPCResult<{ success: boolean; error?: string }>> => {
-      try {
-        const result = await startFalkorDB(port);
-        return { success: true, data: result };
-      } catch (error) {
-        return {
-          success: false,
-          error: error instanceof Error ? error.message : 'Failed to start FalkorDB',
-        };
-      }
-    }
-  );
-
-  // Stop FalkorDB container
-  ipcMain.handle(
-    IPC_CHANNELS.DOCKER_STOP_FALKORDB,
-    async (): Promise<IPCResult<{ success: boolean; error?: string }>> => {
-      try {
-        const result = await stopFalkorDB();
-        return { success: true, data: result };
-      } catch (error) {
-        return {
-          success: false,
-          error: error instanceof Error ? error.message : 'Failed to stop FalkorDB',
-        };
-      }
-    }
-  );
-
-  // Open Docker Desktop application
-  ipcMain.handle(
-    IPC_CHANNELS.DOCKER_OPEN_DESKTOP,
-    async (): Promise<IPCResult<{ success: boolean; error?: string }>> => {
-      try {
-        const result = await openDockerDesktop();
-        return { success: true, data: result };
-      } catch (error) {
-        return {
-          success: false,
-          error: error instanceof Error ? error.message : 'Failed to open Docker Desktop',
-        };
-      }
-    }
-  );
-
-  // Get Docker download URL
-  ipcMain.handle(IPC_CHANNELS.DOCKER_GET_DOWNLOAD_URL, async (): Promise<string> => {
-    return getDockerDownloadUrl();
-  });
-
-  // ============================================
-  // Graphiti Validation Handlers
-  // ============================================
-
-  // Validate FalkorDB connection
-  ipcMain.handle(
-    IPC_CHANNELS.GRAPHITI_VALIDATE_FALKORDB,
-    async (_, uri: string): Promise<IPCResult<GraphitiValidationResult>> => {
-      try {
-        const result = await validateFalkorDBConnection(uri);
-        return { success: true, data: result };
-      } catch (error) {
-        return {
-          success: false,
-          error: error instanceof Error ? error.message : 'Failed to validate FalkorDB connection',
-        };
-      }
-    }
-  );
-
-  // Validate OpenAI API key
-  ipcMain.handle(
-    IPC_CHANNELS.GRAPHITI_VALIDATE_OPENAI,
-    async (_, apiKey: string): Promise<IPCResult<GraphitiValidationResult>> => {
-      try {
-        const result = await validateOpenAIApiKey(apiKey);
-        return { success: true, data: result };
-      } catch (error) {
-        return {
-          success: false,
-          error: error instanceof Error ? error.message : 'Failed to validate OpenAI API key',
-        };
-      }
-    }
-  );
-
-  // Test full Graphiti connection (FalkorDB + OpenAI)
-  ipcMain.handle(
-    IPC_CHANNELS.GRAPHITI_TEST_CONNECTION,
-    async (
-      _,
-      falkorDbUri: string,
-      openAiApiKey: string
-    ): Promise<IPCResult<{
-      falkordb: GraphitiValidationResult;
-      openai: GraphitiValidationResult;
-      ready: boolean;
-    }>> => {
-      try {
-        const result = await testGraphitiConnection(falkorDbUri, openAiApiKey);
-        return { success: true, data: result };
-      } catch (error) {
-        return {
-          success: false,
-          error: error instanceof Error ? error.message : 'Failed to test Graphiti connection',
-        };
-      }
-    }
-  );
+  // Register the new memory handlers instead
+  registerMemoryHandlers();
 }

auto-claude-ui/src/main/ipc-handlers/env-handlers.ts

@@ -62,24 +62,49 @@ export function registerEnvHandlers(
     if (config.githubAutoSync !== undefined) {
       existingVars['GITHUB_AUTO_SYNC'] = config.githubAutoSync ? 'true' : 'false';
     }
+    // Git/Worktree Settings
+    if (config.defaultBranch !== undefined) {
+      existingVars['DEFAULT_BRANCH'] = config.defaultBranch;
+    }
     if (config.graphitiEnabled !== undefined) {
       existingVars['GRAPHITI_ENABLED'] = config.graphitiEnabled ? 'true' : 'false';
     }
+    // Memory Provider Configuration (embeddings only - LLM uses Claude SDK)
+    if (config.graphitiProviderConfig) {
+      const pc = config.graphitiProviderConfig;
+      // Embedding provider only (LLM provider removed - Claude SDK handles RAG)
+      if (pc.embeddingProvider) existingVars['GRAPHITI_EMBEDDER_PROVIDER'] = pc.embeddingProvider;
+      // OpenAI Embeddings
+      if (pc.openaiApiKey) existingVars['OPENAI_API_KEY'] = pc.openaiApiKey;
+      if (pc.openaiEmbeddingModel) existingVars['OPENAI_EMBEDDING_MODEL'] = pc.openaiEmbeddingModel;
+      // Azure OpenAI Embeddings
+      if (pc.azureOpenaiApiKey) existingVars['AZURE_OPENAI_API_KEY'] = pc.azureOpenaiApiKey;
+      if (pc.azureOpenaiBaseUrl) existingVars['AZURE_OPENAI_BASE_URL'] = pc.azureOpenaiBaseUrl;
+      if (pc.azureOpenaiEmbeddingDeployment) existingVars['AZURE_OPENAI_EMBEDDING_DEPLOYMENT'] = pc.azureOpenaiEmbeddingDeployment;
+      // Voyage Embeddings
+      if (pc.voyageApiKey) existingVars['VOYAGE_API_KEY'] = pc.voyageApiKey;
+      if (pc.voyageEmbeddingModel) existingVars['VOYAGE_EMBEDDING_MODEL'] = pc.voyageEmbeddingModel;
+      // Google Embeddings
+      if (pc.googleApiKey) existingVars['GOOGLE_API_KEY'] = pc.googleApiKey;
+      if (pc.googleEmbeddingModel) existingVars['GOOGLE_EMBEDDING_MODEL'] = pc.googleEmbeddingModel;
+      // Ollama Embeddings
+      if (pc.ollamaBaseUrl) existingVars['OLLAMA_BASE_URL'] = pc.ollamaBaseUrl;
+      if (pc.ollamaEmbeddingModel) existingVars['OLLAMA_EMBEDDING_MODEL'] = pc.ollamaEmbeddingModel;
+      if (pc.ollamaEmbeddingDim) existingVars['OLLAMA_EMBEDDING_DIM'] = String(pc.ollamaEmbeddingDim);
+      // LadybugDB (embedded database)
+      if (pc.dbPath) existingVars['GRAPHITI_DB_PATH'] = pc.dbPath;
+      if (pc.database) existingVars['GRAPHITI_DATABASE'] = pc.database;
+    }
+    // Legacy fields (still supported)
     if (config.openaiApiKey !== undefined) {
       existingVars['OPENAI_API_KEY'] = config.openaiApiKey;
     }
-    if (config.graphitiFalkorDbHost !== undefined) {
-      existingVars['GRAPHITI_FALKORDB_HOST'] = config.graphitiFalkorDbHost;
-    }
-    if (config.graphitiFalkorDbPort !== undefined) {
-      existingVars['GRAPHITI_FALKORDB_PORT'] = String(config.graphitiFalkorDbPort);
-    }
-    if (config.graphitiFalkorDbPassword !== undefined) {
-      existingVars['GRAPHITI_FALKORDB_PASSWORD'] = config.graphitiFalkorDbPassword;
-    }
     if (config.graphitiDatabase !== undefined) {
       existingVars['GRAPHITI_DATABASE'] = config.graphitiDatabase;
     }
+    if (config.graphitiDbPath !== undefined) {
+      existingVars['GRAPHITI_DB_PATH'] = config.graphitiDbPath;
+    }
     if (config.enableFancyUi !== undefined) {
       existingVars['ENABLE_FANCY_UI'] = config.enableFancyUi ? 'true' : 'false';
     }
@@ -109,20 +134,52 @@ ${existingVars['GITHUB_TOKEN'] ? `GITHUB_TOKEN=${existingVars['GITHUB_TOKEN']}`
 ${existingVars['GITHUB_REPO'] ? `GITHUB_REPO=${existingVars['GITHUB_REPO']}` : '# GITHUB_REPO=owner/repo'}
 ${existingVars['GITHUB_AUTO_SYNC'] !== undefined ? `GITHUB_AUTO_SYNC=${existingVars['GITHUB_AUTO_SYNC']}` : '# GITHUB_AUTO_SYNC=false'}
 
+# =============================================================================
+# GIT/WORKTREE SETTINGS (OPTIONAL)
+# =============================================================================
+# Default base branch for worktree creation
+# If not set, Auto Claude will auto-detect main/master, or fall back to current branch
+${existingVars['DEFAULT_BRANCH'] ? `DEFAULT_BRANCH=${existingVars['DEFAULT_BRANCH']}` : '# DEFAULT_BRANCH=main'}
+
 # =============================================================================
 # UI SETTINGS (OPTIONAL)
 # =============================================================================
 ${existingVars['ENABLE_FANCY_UI'] !== undefined ? `ENABLE_FANCY_UI=${existingVars['ENABLE_FANCY_UI']}` : '# ENABLE_FANCY_UI=true'}
 
 # =============================================================================
-# GRAPHITI MEMORY INTEGRATION (OPTIONAL)
+# MEMORY INTEGRATION
+# Embedding providers: OpenAI, Google AI, Azure OpenAI, Ollama, Voyage
 # =============================================================================
-${existingVars['GRAPHITI_ENABLED'] ? `GRAPHITI_ENABLED=${existingVars['GRAPHITI_ENABLED']}` : '# GRAPHITI_ENABLED=false'}
+${existingVars['GRAPHITI_ENABLED'] ? `GRAPHITI_ENABLED=${existingVars['GRAPHITI_ENABLED']}` : '# GRAPHITI_ENABLED=true'}
+
+# Embedding Provider (for semantic search - optional, keyword search works without)
+${existingVars['GRAPHITI_EMBEDDER_PROVIDER'] ? `GRAPHITI_EMBEDDER_PROVIDER=${existingVars['GRAPHITI_EMBEDDER_PROVIDER']}` : '# GRAPHITI_EMBEDDER_PROVIDER=ollama'}
+
+# OpenAI Embeddings
 ${existingVars['OPENAI_API_KEY'] ? `OPENAI_API_KEY=${existingVars['OPENAI_API_KEY']}` : '# OPENAI_API_KEY='}
-${existingVars['GRAPHITI_FALKORDB_HOST'] ? `GRAPHITI_FALKORDB_HOST=${existingVars['GRAPHITI_FALKORDB_HOST']}` : '# GRAPHITI_FALKORDB_HOST=localhost'}
-${existingVars['GRAPHITI_FALKORDB_PORT'] ? `GRAPHITI_FALKORDB_PORT=${existingVars['GRAPHITI_FALKORDB_PORT']}` : '# GRAPHITI_FALKORDB_PORT=6380'}
-${existingVars['GRAPHITI_FALKORDB_PASSWORD'] ? `GRAPHITI_FALKORDB_PASSWORD=${existingVars['GRAPHITI_FALKORDB_PASSWORD']}` : '# GRAPHITI_FALKORDB_PASSWORD='}
+${existingVars['OPENAI_EMBEDDING_MODEL'] ? `OPENAI_EMBEDDING_MODEL=${existingVars['OPENAI_EMBEDDING_MODEL']}` : '# OPENAI_EMBEDDING_MODEL=text-embedding-3-small'}
+
+# Azure OpenAI Embeddings
+${existingVars['AZURE_OPENAI_API_KEY'] ? `AZURE_OPENAI_API_KEY=${existingVars['AZURE_OPENAI_API_KEY']}` : '# AZURE_OPENAI_API_KEY='}
+${existingVars['AZURE_OPENAI_BASE_URL'] ? `AZURE_OPENAI_BASE_URL=${existingVars['AZURE_OPENAI_BASE_URL']}` : '# AZURE_OPENAI_BASE_URL='}
+${existingVars['AZURE_OPENAI_EMBEDDING_DEPLOYMENT'] ? `AZURE_OPENAI_EMBEDDING_DEPLOYMENT=${existingVars['AZURE_OPENAI_EMBEDDING_DEPLOYMENT']}` : '# AZURE_OPENAI_EMBEDDING_DEPLOYMENT='}
+
+# Voyage AI Embeddings
+${existingVars['VOYAGE_API_KEY'] ? `VOYAGE_API_KEY=${existingVars['VOYAGE_API_KEY']}` : '# VOYAGE_API_KEY='}
+${existingVars['VOYAGE_EMBEDDING_MODEL'] ? `VOYAGE_EMBEDDING_MODEL=${existingVars['VOYAGE_EMBEDDING_MODEL']}` : '# VOYAGE_EMBEDDING_MODEL=voyage-3'}
+
+# Google AI Embeddings
+${existingVars['GOOGLE_API_KEY'] ? `GOOGLE_API_KEY=${existingVars['GOOGLE_API_KEY']}` : '# GOOGLE_API_KEY='}
+${existingVars['GOOGLE_EMBEDDING_MODEL'] ? `GOOGLE_EMBEDDING_MODEL=${existingVars['GOOGLE_EMBEDDING_MODEL']}` : '# GOOGLE_EMBEDDING_MODEL=text-embedding-004'}
+
+# Ollama Embeddings (Local - free)
+${existingVars['OLLAMA_BASE_URL'] ? `OLLAMA_BASE_URL=${existingVars['OLLAMA_BASE_URL']}` : '# OLLAMA_BASE_URL=http://localhost:11434'}
+${existingVars['OLLAMA_EMBEDDING_MODEL'] ? `OLLAMA_EMBEDDING_MODEL=${existingVars['OLLAMA_EMBEDDING_MODEL']}` : '# OLLAMA_EMBEDDING_MODEL=embeddinggemma'}
+${existingVars['OLLAMA_EMBEDDING_DIM'] ? `OLLAMA_EMBEDDING_DIM=${existingVars['OLLAMA_EMBEDDING_DIM']}` : '# OLLAMA_EMBEDDING_DIM=768'}
+
+# LadybugDB Database (embedded - no Docker required)
 ${existingVars['GRAPHITI_DATABASE'] ? `GRAPHITI_DATABASE=${existingVars['GRAPHITI_DATABASE']}` : '# GRAPHITI_DATABASE=auto_claude_memory'}
+${existingVars['GRAPHITI_DB_PATH'] ? `GRAPHITI_DB_PATH=${existingVars['GRAPHITI_DB_PATH']}` : '# GRAPHITI_DB_PATH=~/.auto-claude/memories'}
 `;
 
     return content;
@@ -216,6 +273,11 @@ ${existingVars['GRAPHITI_DATABASE'] ? `GRAPHITI_DATABASE=${existingVars['GRAPHIT
         config.githubAutoSync = true;
       }
 
+      // Git/Worktree config
+      if (vars['DEFAULT_BRANCH']) {
+        config.defaultBranch = vars['DEFAULT_BRANCH'];
+      }
+
       if (vars['GRAPHITI_ENABLED']?.toLowerCase() === 'true') {
         config.graphitiEnabled = true;
       }
@@ -229,23 +291,46 @@ ${existingVars['GRAPHITI_DATABASE'] ? `GRAPHITI_DATABASE=${existingVars['GRAPHIT
         config.openaiKeyIsGlobal = true;
       }
 
-      if (vars['GRAPHITI_FALKORDB_HOST']) {
-        config.graphitiFalkorDbHost = vars['GRAPHITI_FALKORDB_HOST'];
-      }
-      if (vars['GRAPHITI_FALKORDB_PORT']) {
-        config.graphitiFalkorDbPort = parseInt(vars['GRAPHITI_FALKORDB_PORT'], 10);
-      }
-      if (vars['GRAPHITI_FALKORDB_PASSWORD']) {
-        config.graphitiFalkorDbPassword = vars['GRAPHITI_FALKORDB_PASSWORD'];
-      }
       if (vars['GRAPHITI_DATABASE']) {
         config.graphitiDatabase = vars['GRAPHITI_DATABASE'];
       }
+      if (vars['GRAPHITI_DB_PATH']) {
+        config.graphitiDbPath = vars['GRAPHITI_DB_PATH'];
+      }
 
       if (vars['ENABLE_FANCY_UI']?.toLowerCase() === 'false') {
         config.enableFancyUi = false;
       }
 
+      // Populate graphitiProviderConfig from .env file (embeddings only - no LLM provider)
+      const embeddingProvider = vars['GRAPHITI_EMBEDDER_PROVIDER'];
+      if (embeddingProvider || vars['AZURE_OPENAI_API_KEY'] ||
+          vars['VOYAGE_API_KEY'] || vars['GOOGLE_API_KEY'] || vars['OLLAMA_BASE_URL']) {
+        config.graphitiProviderConfig = {
+          embeddingProvider: (embeddingProvider as 'openai' | 'voyage' | 'azure_openai' | 'ollama' | 'google') || 'ollama',
+          // OpenAI Embeddings
+          openaiApiKey: vars['OPENAI_API_KEY'],
+          openaiEmbeddingModel: vars['OPENAI_EMBEDDING_MODEL'],
+          // Azure OpenAI Embeddings
+          azureOpenaiApiKey: vars['AZURE_OPENAI_API_KEY'],
+          azureOpenaiBaseUrl: vars['AZURE_OPENAI_BASE_URL'],
+          azureOpenaiEmbeddingDeployment: vars['AZURE_OPENAI_EMBEDDING_DEPLOYMENT'],
+          // Voyage Embeddings
+          voyageApiKey: vars['VOYAGE_API_KEY'],
+          voyageEmbeddingModel: vars['VOYAGE_EMBEDDING_MODEL'],
+          // Google Embeddings
+          googleApiKey: vars['GOOGLE_API_KEY'],
+          googleEmbeddingModel: vars['GOOGLE_EMBEDDING_MODEL'],
+          // Ollama Embeddings
+          ollamaBaseUrl: vars['OLLAMA_BASE_URL'],
+          ollamaEmbeddingModel: vars['OLLAMA_EMBEDDING_MODEL'],
+          ollamaEmbeddingDim: vars['OLLAMA_EMBEDDING_DIM'] ? parseInt(vars['OLLAMA_EMBEDDING_DIM'], 10) : undefined,
+          // LadybugDB
+          database: vars['GRAPHITI_DATABASE'],
+          dbPath: vars['GRAPHITI_DB_PATH'],
+        };
+      }
+
       return { success: true, data: config };
     }
   );

auto-claude-ui/src/main/ipc-handlers/file-handlers.ts

@@ -8,8 +8,8 @@ import type { IPCResult, FileNode } from '../../shared/types';
 const IGNORED_DIRS = new Set([
   'node_modules', '.git', '__pycache__', 'dist', 'build',
   '.next', '.nuxt', 'coverage', '.cache', '.venv', 'venv',
-  '.idea', '.vscode', 'out', '.turbo', '.auto-claude',
-  '.worktrees', 'vendor', 'target', '.gradle', '.maven'
+  'out', '.turbo', '.worktrees',
+  'vendor', 'target', '.gradle', '.maven'
 ]);
 
 /**
@@ -29,8 +29,11 @@ export function registerFileHandlers(): void {
         // Filter and map entries
         const nodes: FileNode[] = [];
         for (const entry of entries) {
-          // Skip hidden files (except .env which is often useful)
-          if (entry.name.startsWith('.') && entry.name !== '.env') continue;
+          // Skip hidden files (not directories) except useful ones like .env, .gitignore
+          if (!entry.isDirectory() && entry.name.startsWith('.') &&
+              !['.env', '.gitignore', '.env.example', '.env.local'].includes(entry.name)) {
+            continue;
+          }
           // Skip ignored directories
           if (entry.isDirectory() && IGNORED_DIRS.has(entry.name)) continue;
 

auto-claude-ui/src/main/ipc-handlers/github/__tests__/oauth-handlers.spec.ts

@@ -0,0 +1,548 @@
+/**
+ * Unit tests for GitHub OAuth handlers
+ * Tests device code parsing, shell.openExternal handling, and error recovery
+ */
+import { describe, it, expect, vi, beforeEach, afterEach } from 'vitest';
+import { EventEmitter } from 'events';
+
+// Mock child_process before importing
+const mockSpawn = vi.fn();
+const mockExecSync = vi.fn();
+const mockExecFileSync = vi.fn();
+
+vi.mock('child_process', () => ({
+  spawn: (...args: unknown[]) => mockSpawn(...args),
+  execSync: (...args: unknown[]) => mockExecSync(...args),
+  execFileSync: (...args: unknown[]) => mockExecFileSync(...args)
+}));
+
+// Mock shell.openExternal
+const mockOpenExternal = vi.fn();
+
+vi.mock('electron', () => {
+  const mockIpcMain = new (class extends EventEmitter {
+    private handlers: Map<string, Function> = new Map();
+
+    handle(channel: string, handler: Function): void {
+      this.handlers.set(channel, handler);
+    }
+
+    removeHandler(channel: string): void {
+      this.handlers.delete(channel);
+    }
+
+    async invokeHandler(channel: string, event: unknown, ...args: unknown[]): Promise<unknown> {
+      const handler = this.handlers.get(channel);
+      if (handler) {
+        return handler(event, ...args);
+      }
+      throw new Error(`No handler for channel: ${channel}`);
+    }
+
+    getHandler(channel: string): Function | undefined {
+      return this.handlers.get(channel);
+    }
+  })();
+
+  return {
+    ipcMain: mockIpcMain,
+    shell: {
+      openExternal: (...args: unknown[]) => mockOpenExternal(...args)
+    }
+  };
+});
+
+// Mock @electron-toolkit/utils
+vi.mock('@electron-toolkit/utils', () => ({
+  is: {
+    dev: true,
+    windows: process.platform === 'win32',
+    macos: process.platform === 'darwin',
+    linux: process.platform === 'linux'
+  }
+}));
+
+// Create mock process for spawn
+function createMockProcess(): EventEmitter & {
+  stdout: EventEmitter | null;
+  stderr: EventEmitter | null;
+  stdin: { write: ReturnType<typeof vi.fn>; end: ReturnType<typeof vi.fn> } | null;
+} {
+  const proc = new EventEmitter() as EventEmitter & {
+    stdout: EventEmitter | null;
+    stderr: EventEmitter | null;
+    stdin: { write: ReturnType<typeof vi.fn>; end: ReturnType<typeof vi.fn> } | null;
+  };
+  proc.stdout = new EventEmitter();
+  proc.stderr = new EventEmitter();
+  proc.stdin = { write: vi.fn(), end: vi.fn() };
+  return proc;
+}
+
+describe('GitHub OAuth Handlers', () => {
+  let ipcMain: EventEmitter & {
+    handlers: Map<string, Function>;
+    invokeHandler: (channel: string, event: unknown, ...args: unknown[]) => Promise<unknown>;
+    getHandler: (channel: string) => Function | undefined;
+  };
+
+  beforeEach(async () => {
+    vi.clearAllMocks();
+    vi.resetModules();
+
+    // Get mocked ipcMain
+    const electron = await import('electron');
+    ipcMain = electron.ipcMain as unknown as typeof ipcMain;
+  });
+
+  afterEach(() => {
+    vi.clearAllMocks();
+  });
+
+  describe('Device Code Parsing', () => {
+    it('should parse device code from standard gh CLI output format', async () => {
+      const mockProcess = createMockProcess();
+      mockSpawn.mockReturnValue(mockProcess);
+      mockOpenExternal.mockResolvedValue(undefined);
+
+      const { registerStartGhAuth } = await import('../oauth-handlers');
+      registerStartGhAuth();
+
+      // Start the handler
+      const resultPromise = ipcMain.invokeHandler('github:startAuth', {});
+
+      // Simulate gh CLI output with device code
+      mockProcess.stderr?.emit('data', '! First copy your one-time code: ABCD-1234\n');
+      mockProcess.stderr?.emit('data', '- Press Enter to open github.com in your browser...\n');
+
+      // Complete the process
+      mockProcess.emit('close', 0);
+
+      const result = await resultPromise;
+
+      expect(result).toHaveProperty('success', true);
+      expect(result).toHaveProperty('data');
+      const data = (result as { data: { deviceCode: string } }).data;
+      expect(data.deviceCode).toBe('ABCD-1234');
+    });
+
+    it('should parse device code from alternate output format (lowercase "code")', async () => {
+      const mockProcess = createMockProcess();
+      mockSpawn.mockReturnValue(mockProcess);
+      mockOpenExternal.mockResolvedValue(undefined);
+
+      const { registerStartGhAuth } = await import('../oauth-handlers');
+      registerStartGhAuth();
+
+      const resultPromise = ipcMain.invokeHandler('github:startAuth', {});
+
+      // Alternate format: "code: XXXX-XXXX" without "one-time"
+      mockProcess.stderr?.emit('data', 'Enter the code: EFGH-5678\n');
+      mockProcess.emit('close', 0);
+
+      const result = await resultPromise;
+
+      expect(result).toHaveProperty('success', true);
+      const data = (result as { data: { deviceCode: string } }).data;
+      expect(data.deviceCode).toBe('EFGH-5678');
+    });
+
+    it('should parse device code from stdout (not just stderr)', async () => {
+      const mockProcess = createMockProcess();
+      mockSpawn.mockReturnValue(mockProcess);
+      mockOpenExternal.mockResolvedValue(undefined);
+
+      const { registerStartGhAuth } = await import('../oauth-handlers');
+      registerStartGhAuth();
+
+      const resultPromise = ipcMain.invokeHandler('github:startAuth', {});
+
+      // Device code in stdout instead of stderr
+      mockProcess.stdout?.emit('data', '! First copy your one-time code: IJKL-9012\n');
+      mockProcess.emit('close', 0);
+
+      const result = await resultPromise;
+
+      expect(result).toHaveProperty('success', true);
+      const data = (result as { data: { deviceCode: string } }).data;
+      expect(data.deviceCode).toBe('IJKL-9012');
+    });
+
+    it('should handle output without device code gracefully', async () => {
+      const mockProcess = createMockProcess();
+      mockSpawn.mockReturnValue(mockProcess);
+
+      const { registerStartGhAuth } = await import('../oauth-handlers');
+      registerStartGhAuth();
+
+      const resultPromise = ipcMain.invokeHandler('github:startAuth', {});
+
+      // Output without device code
+      mockProcess.stderr?.emit('data', 'Some other message\n');
+      mockProcess.emit('close', 0);
+
+      const result = await resultPromise;
+
+      expect(result).toHaveProperty('success', true);
+      const data = (result as { data: { deviceCode?: string } }).data;
+      expect(data.deviceCode).toBeUndefined();
+    });
+
+    it('should extract URL from output containing https://github.com/login/device', async () => {
+      const mockProcess = createMockProcess();
+      mockSpawn.mockReturnValue(mockProcess);
+      mockOpenExternal.mockResolvedValue(undefined);
+
+      const { registerStartGhAuth } = await import('../oauth-handlers');
+      registerStartGhAuth();
+
+      const resultPromise = ipcMain.invokeHandler('github:startAuth', {});
+
+      mockProcess.stderr?.emit('data', '! First copy your one-time code: MNOP-3456\n');
+      mockProcess.stderr?.emit('data', 'Then visit https://github.com/login/device to authenticate\n');
+      mockProcess.emit('close', 0);
+
+      const result = await resultPromise;
+
+      expect(result).toHaveProperty('success', true);
+      const data = (result as { data: { authUrl: string } }).data;
+      expect(data.authUrl).toBe('https://github.com/login/device');
+    });
+  });
+
+  describe('shell.openExternal Handling', () => {
+    it('should call shell.openExternal with extracted URL when device code found', async () => {
+      const mockProcess = createMockProcess();
+      mockSpawn.mockReturnValue(mockProcess);
+      mockOpenExternal.mockResolvedValue(undefined);
+
+      const { registerStartGhAuth } = await import('../oauth-handlers');
+      registerStartGhAuth();
+
+      const resultPromise = ipcMain.invokeHandler('github:startAuth', {});
+
+      mockProcess.stderr?.emit('data', '! First copy your one-time code: QRST-7890\n');
+
+      // Wait for next tick to allow async browser opening
+      await new Promise(resolve => setTimeout(resolve, 10));
+
+      mockProcess.emit('close', 0);
+      await resultPromise;
+
+      expect(mockOpenExternal).toHaveBeenCalledWith('https://github.com/login/device');
+    });
+
+    it('should set browserOpened to true when shell.openExternal succeeds', async () => {
+      const mockProcess = createMockProcess();
+      mockSpawn.mockReturnValue(mockProcess);
+      mockOpenExternal.mockResolvedValue(undefined);
+
+      const { registerStartGhAuth } = await import('../oauth-handlers');
+      registerStartGhAuth();
+
+      const resultPromise = ipcMain.invokeHandler('github:startAuth', {});
+
+      mockProcess.stderr?.emit('data', '! First copy your one-time code: UVWX-1234\n');
+
+      // Wait for async browser opening
+      await new Promise(resolve => setTimeout(resolve, 10));
+
+      mockProcess.emit('close', 0);
+      const result = await resultPromise;
+
+      expect(result).toHaveProperty('success', true);
+      const data = (result as { data: { browserOpened: boolean } }).data;
+      expect(data.browserOpened).toBe(true);
+    });
+
+    it('should set browserOpened to false when shell.openExternal fails', async () => {
+      const mockProcess = createMockProcess();
+      mockSpawn.mockReturnValue(mockProcess);
+      mockOpenExternal.mockRejectedValue(new Error('Failed to open browser'));
+
+      const { registerStartGhAuth } = await import('../oauth-handlers');
+      registerStartGhAuth();
+
+      const resultPromise = ipcMain.invokeHandler('github:startAuth', {});
+
+      mockProcess.stderr?.emit('data', '! First copy your one-time code: YZAB-5678\n');
+
+      // Wait for async browser opening to fail
+      await new Promise(resolve => setTimeout(resolve, 10));
+
+      mockProcess.emit('close', 0);
+      const result = await resultPromise;
+
+      expect(result).toHaveProperty('success', true);
+      const data = (result as { data: { browserOpened: boolean } }).data;
+      expect(data.browserOpened).toBe(false);
+    });
+
+    it('should provide fallbackUrl when browser fails to open', async () => {
+      const mockProcess = createMockProcess();
+      mockSpawn.mockReturnValue(mockProcess);
+      mockOpenExternal.mockRejectedValue(new Error('Failed to open browser'));
+
+      const { registerStartGhAuth } = await import('../oauth-handlers');
+      registerStartGhAuth();
+
+      const resultPromise = ipcMain.invokeHandler('github:startAuth', {});
+
+      mockProcess.stderr?.emit('data', '! First copy your one-time code: CDEF-9012\n');
+
+      // Wait for async browser opening to fail
+      await new Promise(resolve => setTimeout(resolve, 10));
+
+      mockProcess.emit('close', 0);
+      const result = await resultPromise;
+
+      expect(result).toHaveProperty('success', true);
+      const data = (result as { data: { fallbackUrl?: string } }).data;
+      expect(data.fallbackUrl).toBe('https://github.com/login/device');
+    });
+
+    it('should not provide fallbackUrl when browser opens successfully', async () => {
+      const mockProcess = createMockProcess();
+      mockSpawn.mockReturnValue(mockProcess);
+      mockOpenExternal.mockResolvedValue(undefined);
+
+      const { registerStartGhAuth } = await import('../oauth-handlers');
+      registerStartGhAuth();
+
+      const resultPromise = ipcMain.invokeHandler('github:startAuth', {});
+
+      mockProcess.stderr?.emit('data', '! First copy your one-time code: GHIJ-3456\n');
+
+      // Wait for async browser opening
+      await new Promise(resolve => setTimeout(resolve, 10));
+
+      mockProcess.emit('close', 0);
+      const result = await resultPromise;
+
+      expect(result).toHaveProperty('success', true);
+      const data = (result as { data: { fallbackUrl?: string } }).data;
+      expect(data.fallbackUrl).toBeUndefined();
+    });
+  });
+
+  describe('Error Handling', () => {
+    it('should handle gh CLI process error', async () => {
+      const mockProcess = createMockProcess();
+      mockSpawn.mockReturnValue(mockProcess);
+
+      const { registerStartGhAuth } = await import('../oauth-handlers');
+      registerStartGhAuth();
+
+      const resultPromise = ipcMain.invokeHandler('github:startAuth', {});
+
+      // Emit error event
+      mockProcess.emit('error', new Error('spawn gh ENOENT'));
+
+      const result = await resultPromise;
+
+      expect(result).toHaveProperty('success', false);
+      expect(result).toHaveProperty('error', 'spawn gh ENOENT');
+      const data = (result as { data: { fallbackUrl: string } }).data;
+      expect(data.fallbackUrl).toBe('https://github.com/login/device');
+    });
+
+    it('should handle non-zero exit code', async () => {
+      const mockProcess = createMockProcess();
+      mockSpawn.mockReturnValue(mockProcess);
+
+      const { registerStartGhAuth } = await import('../oauth-handlers');
+      registerStartGhAuth();
+
+      const resultPromise = ipcMain.invokeHandler('github:startAuth', {});
+
+      mockProcess.stderr?.emit('data', 'error: some authentication error\n');
+      mockProcess.emit('close', 1);
+
+      const result = await resultPromise;
+
+      expect(result).toHaveProperty('success', false);
+      const data = (result as { data: { fallbackUrl: string } }).data;
+      expect(data.fallbackUrl).toBe('https://github.com/login/device');
+    });
+
+    it('should include device code in error result if it was extracted before failure', async () => {
+      const mockProcess = createMockProcess();
+      mockSpawn.mockReturnValue(mockProcess);
+      mockOpenExternal.mockResolvedValue(undefined);
+
+      const { registerStartGhAuth } = await import('../oauth-handlers');
+      registerStartGhAuth();
+
+      const resultPromise = ipcMain.invokeHandler('github:startAuth', {});
+
+      // Device code output followed by failure
+      mockProcess.stderr?.emit('data', '! First copy your one-time code: KLMN-7890\n');
+
+      // Wait for async browser opening
+      await new Promise(resolve => setTimeout(resolve, 10));
+
+      mockProcess.stderr?.emit('data', 'error: authentication failed\n');
+      mockProcess.emit('close', 1);
+
+      const result = await resultPromise;
+
+      expect(result).toHaveProperty('success', false);
+      const data = (result as { data: { deviceCode: string; fallbackUrl: string } }).data;
+      expect(data.deviceCode).toBe('KLMN-7890');
+      expect(data.fallbackUrl).toBe('https://github.com/login/device');
+    });
+
+    it('should provide user-friendly error message on process spawn failure', async () => {
+      const mockProcess = createMockProcess();
+      mockSpawn.mockReturnValue(mockProcess);
+
+      const { registerStartGhAuth } = await import('../oauth-handlers');
+      registerStartGhAuth();
+
+      const resultPromise = ipcMain.invokeHandler('github:startAuth', {});
+
+      mockProcess.emit('error', new Error('spawn gh ENOENT'));
+
+      const result = await resultPromise;
+
+      expect(result).toHaveProperty('success', false);
+      const data = (result as { data: { message: string } }).data;
+      expect(data.message).toContain('Failed to start GitHub CLI');
+    });
+  });
+
+  describe('gh CLI Check Handler', () => {
+    it('should return installed: true when gh CLI is found', async () => {
+      mockExecSync.mockImplementation((cmd: string) => {
+        if (cmd.includes('which gh') || cmd.includes('where gh')) {
+          return '/usr/local/bin/gh\n';
+        }
+        if (cmd === 'gh --version') {
+          return 'gh version 2.65.0 (2024-01-15)\n';
+        }
+        return '';
+      });
+
+      const { registerCheckGhCli } = await import('../oauth-handlers');
+      registerCheckGhCli();
+
+      const result = await ipcMain.invokeHandler('github:checkCli', {});
+
+      expect(result).toHaveProperty('success', true);
+      const data = (result as { data: { installed: boolean; version: string } }).data;
+      expect(data.installed).toBe(true);
+      expect(data.version).toContain('gh version');
+    });
+
+    it('should return installed: false when gh CLI is not found', async () => {
+      mockExecSync.mockImplementation(() => {
+        throw new Error('Command not found');
+      });
+
+      const { registerCheckGhCli } = await import('../oauth-handlers');
+      registerCheckGhCli();
+
+      const result = await ipcMain.invokeHandler('github:checkCli', {});
+
+      expect(result).toHaveProperty('success', true);
+      const data = (result as { data: { installed: boolean } }).data;
+      expect(data.installed).toBe(false);
+    });
+  });
+
+  describe('gh Auth Check Handler', () => {
+    it('should return authenticated: true with username when logged in', async () => {
+      mockExecSync.mockImplementation((cmd: string) => {
+        if (cmd === 'gh auth status') {
+          return 'Logged in to github.com as testuser\n';
+        }
+        if (cmd === 'gh api user --jq .login') {
+          return 'testuser\n';
+        }
+        return '';
+      });
+
+      const { registerCheckGhAuth } = await import('../oauth-handlers');
+      registerCheckGhAuth();
+
+      const result = await ipcMain.invokeHandler('github:checkAuth', {});
+
+      expect(result).toHaveProperty('success', true);
+      const data = (result as { data: { authenticated: boolean; username: string } }).data;
+      expect(data.authenticated).toBe(true);
+      expect(data.username).toBe('testuser');
+    });
+
+    it('should return authenticated: false when not logged in', async () => {
+      mockExecSync.mockImplementation(() => {
+        throw new Error('You are not logged into any GitHub hosts');
+      });
+
+      const { registerCheckGhAuth } = await import('../oauth-handlers');
+      registerCheckGhAuth();
+
+      const result = await ipcMain.invokeHandler('github:checkAuth', {});
+
+      expect(result).toHaveProperty('success', true);
+      const data = (result as { data: { authenticated: boolean } }).data;
+      expect(data.authenticated).toBe(false);
+    });
+  });
+
+  describe('Spawn Arguments', () => {
+    it('should spawn gh with correct auth login arguments', async () => {
+      const mockProcess = createMockProcess();
+      mockSpawn.mockReturnValue(mockProcess);
+
+      const { registerStartGhAuth } = await import('../oauth-handlers');
+      registerStartGhAuth();
+
+      ipcMain.invokeHandler('github:startAuth', {});
+
+      expect(mockSpawn).toHaveBeenCalledWith(
+        'gh',
+        ['auth', 'login', '--web', '--scopes', 'repo'],
+        expect.objectContaining({
+          stdio: ['pipe', 'pipe', 'pipe']
+        })
+      );
+    });
+  });
+
+  describe('Repository Validation', () => {
+    it('should reject invalid repository format', async () => {
+      const { registerGetGitHubBranches } = await import('../oauth-handlers');
+      registerGetGitHubBranches();
+
+      // Test with injection attempt
+      const result = await ipcMain.invokeHandler(
+        'github:getBranches',
+        {},
+        'owner/repo; rm -rf /',
+        'token'
+      );
+
+      expect(result).toHaveProperty('success', false);
+      expect(result).toHaveProperty('error', 'Invalid repository format. Expected: owner/repo');
+    });
+
+    it('should accept valid repository format', async () => {
+      mockExecFileSync.mockReturnValue('main\nfeature-branch\n');
+
+      const { registerGetGitHubBranches } = await import('../oauth-handlers');
+      registerGetGitHubBranches();
+
+      const result = await ipcMain.invokeHandler(
+        'github:getBranches',
+        {},
+        'valid-owner/valid-repo',
+        'token'
+      );
+
+      expect(result).toHaveProperty('success', true);
+      const data = (result as { data: string[] }).data;
+      expect(data).toContain('main');
+      expect(data).toContain('feature-branch');
+    });
+  });
+});

auto-claude-ui/src/main/ipc-handlers/github/issue-handlers.ts

@@ -6,7 +6,7 @@ import { ipcMain } from 'electron';
 import { IPC_CHANNELS } from '../../../shared/constants';
 import type { IPCResult, GitHubIssue } from '../../../shared/types';
 import { projectStore } from '../../project-store';
-import { getGitHubConfig, githubFetch } from './utils';
+import { getGitHubConfig, githubFetch, normalizeRepoReference } from './utils';
 import type { GitHubAPIIssue, GitHubAPIComment } from './types';
 
 /**
@@ -57,16 +57,32 @@ export function registerGetIssues(): void {
       }
 
       try {
+        const normalizedRepo = normalizeRepoReference(config.repo);
+        if (!normalizedRepo) {
+          return {
+            success: false,
+            error: 'Invalid repository format. Use owner/repo or GitHub URL.'
+          };
+        }
+
         const issues = await githubFetch(
           config.token,
-          `/repos/${config.repo}/issues?state=${state}&per_page=100&sort=updated`
-        ) as GitHubAPIIssue[];
+          `/repos/${normalizedRepo}/issues?state=${state}&per_page=100&sort=updated`
+        );
+
+        // Ensure issues is an array
+        if (!Array.isArray(issues)) {
+          return {
+            success: false,
+            error: 'Unexpected response format from GitHub API'
+          };
+        }
 
         // Filter out pull requests
-        const issuesOnly = issues.filter(issue => !issue.pull_request);
+        const issuesOnly = issues.filter((issue: GitHubAPIIssue) => !issue.pull_request);
 
-        const result: GitHubIssue[] = issuesOnly.map(issue =>
-          transformIssue(issue, config.repo)
+        const result: GitHubIssue[] = issuesOnly.map((issue: GitHubAPIIssue) =>
+          transformIssue(issue, normalizedRepo)
         );
 
         return { success: true, data: result };
@@ -98,12 +114,20 @@ export function registerGetIssue(): void {
       }
 
       try {
+        const normalizedRepo = normalizeRepoReference(config.repo);
+        if (!normalizedRepo) {
+          return {
+            success: false,
+            error: 'Invalid repository format. Use owner/repo or GitHub URL.'
+          };
+        }
+
         const issue = await githubFetch(
           config.token,
-          `/repos/${config.repo}/issues/${issueNumber}`
+          `/repos/${normalizedRepo}/issues/${issueNumber}`
         ) as GitHubAPIIssue;
 
-        const result = transformIssue(issue, config.repo);
+        const result = transformIssue(issue, normalizedRepo);
 
         return { success: true, data: result };
       } catch (error) {
@@ -134,9 +158,17 @@ export function registerGetIssueComments(): void {
       }
 
       try {
+        const normalizedRepo = normalizeRepoReference(config.repo);
+        if (!normalizedRepo) {
+          return {
+            success: false,
+            error: 'Invalid repository format. Use owner/repo or GitHub URL.'
+          };
+        }
+
         const comments = await githubFetch(
           config.token,
-          `/repos/${config.repo}/issues/${issueNumber}/comments`
+          `/repos/${normalizedRepo}/issues/${issueNumber}/comments`
         ) as GitHubAPIComment[];
 
         return { success: true, data: comments };

auto-claude-ui/src/main/ipc-handlers/github/oauth-handlers.ts

@@ -3,7 +3,7 @@
  * Provides a simpler OAuth flow than manual PAT creation
  */
 
-import { ipcMain } from 'electron';
+import { ipcMain, shell } from 'electron';
 import { execSync, execFileSync, spawn } from 'child_process';
 import { IPC_CHANNELS } from '../../../shared/constants';
 import type { IPCResult } from '../../../shared/types';
@@ -33,6 +33,71 @@ function isValidGitHubRepo(repo: string): boolean {
   return GITHUB_REPO_PATTERN.test(repo);
 }
 
+// Regex patterns for parsing device code from gh CLI output
+// Expected format: "! First copy your one-time code: XXXX-XXXX"
+// Pattern updated to handle different gh CLI versions - supports:
+// - "one-time code", "code", or "verification code" prefixes
+// - Hyphen or space separator in the code (XXXX-XXXX or XXXX XXXX)
+// Note: Separator is REQUIRED to avoid matching 8-char strings without separator
+const DEVICE_CODE_PATTERN = /(?:one-time code|verification code|code):\s*([A-Z0-9]{4}[-\s][A-Z0-9]{4})/i;
+
+// GitHub device flow URL pattern
+const DEVICE_URL_PATTERN = /https:\/\/github\.com\/login\/device/i;
+
+// Default GitHub device flow URL
+const GITHUB_DEVICE_URL = 'https://github.com/login/device';
+
+/**
+ * Parse device code from gh CLI stdout output
+ * Returns the device code (format: XXXX-XXXX) if found, null otherwise
+ * Normalizes space separator to hyphen (GitHub always expects XXXX-XXXX)
+ */
+function parseDeviceCode(output: string): string | null {
+  const match = output.match(DEVICE_CODE_PATTERN);
+  if (match && match[1]) {
+    // Normalize: replace space with hyphen (GitHub expects XXXX-XXXX format)
+    const normalizedCode = match[1].replace(' ', '-');
+    debugLog('Device code extracted successfully (code redacted for security)');
+    return normalizedCode;
+  }
+  return null;
+}
+
+/**
+ * Parse device URL from gh CLI output
+ * Returns the URL if found, or the default GitHub device URL
+ */
+function parseDeviceUrl(output: string): string {
+  const match = output.match(DEVICE_URL_PATTERN);
+  if (match) {
+    debugLog('Found device URL in output:', match[0]);
+    return match[0];
+  }
+  // Default to standard GitHub device flow URL
+  return GITHUB_DEVICE_URL;
+}
+
+/**
+ * Result of parsing device flow output from gh CLI
+ */
+interface DeviceFlowInfo {
+  deviceCode: string | null;
+  authUrl: string;
+}
+
+/**
+ * Parse both device code and URL from combined gh CLI output
+ * Searches through both stdout and stderr as gh may output to either
+ */
+function parseDeviceFlowOutput(stdout: string, stderr: string): DeviceFlowInfo {
+  const combinedOutput = `${stdout}\n${stderr}`;
+
+  return {
+    deviceCode: parseDeviceCode(combinedOutput),
+    authUrl: parseDeviceUrl(combinedOutput)
+  };
+}
+
 /**
  * Check if gh CLI is installed
  */
@@ -114,14 +179,31 @@ export function registerCheckGhAuth(): void {
   );
 }
 
+/**
+ * Result type for GitHub auth start, including device flow information
+ */
+interface GitHubAuthStartResult {
+  success: boolean;
+  message?: string;
+  deviceCode?: string;
+  authUrl?: string;
+  browserOpened?: boolean;
+  /**
+   * Fallback URL provided when browser launch fails.
+   * The frontend should display this URL so users can manually navigate to complete auth.
+   */
+  fallbackUrl?: string;
+}
+
 /**
  * Start GitHub OAuth flow using gh CLI
- * This will open the browser for device flow authentication
+ * This will extract the device code from gh CLI output and open the browser
+ * using Electron's shell.openExternal (bypasses macOS child process restrictions)
  */
 export function registerStartGhAuth(): void {
   ipcMain.handle(
     IPC_CHANNELS.GITHUB_START_AUTH,
-    async (): Promise<IPCResult<{ success: boolean; message?: string }>> => {
+    async (): Promise<IPCResult<GitHubAuthStartResult>> => {
       debugLog('startGitHubAuth handler called');
       return new Promise((resolve) => {
         try {
@@ -135,17 +217,64 @@ export function registerStartGhAuth(): void {
 
           let output = '';
           let errorOutput = '';
+          let deviceCodeExtracted = false;
+          let extractedDeviceCode: string | null = null;
+          let extractedAuthUrl: string = GITHUB_DEVICE_URL;
+          let browserOpenedSuccessfully = false;
+          let extractionInProgress = false;
+
+          // Function to attempt device code extraction and browser opening
+          // Uses mutex pattern to prevent race conditions from concurrent data handlers
+          const tryExtractAndOpenBrowser = async () => {
+            if (deviceCodeExtracted || extractionInProgress) return;
+            extractionInProgress = true;
+
+            const deviceFlowInfo = parseDeviceFlowOutput(output, errorOutput);
+
+            if (deviceFlowInfo.deviceCode) {
+              deviceCodeExtracted = true;
+              extractedDeviceCode = deviceFlowInfo.deviceCode;
+              extractedAuthUrl = deviceFlowInfo.authUrl;
+
+              debugLog('Device code extracted successfully (code redacted for security)');
+              debugLog('Auth URL:', extractedAuthUrl);
+
+              // Open browser using Electron's shell.openExternal
+              // This bypasses macOS child process restrictions that block gh CLI's browser launch
+              try {
+                await shell.openExternal(extractedAuthUrl);
+                browserOpenedSuccessfully = true;
+                debugLog('Browser opened successfully via shell.openExternal');
+              } catch (browserError) {
+                debugLog('Failed to open browser:', browserError instanceof Error ? browserError.message : browserError);
+                browserOpenedSuccessfully = false;
+                // Don't fail here - we'll return the device code so user can manually navigate
+              }
+
+              // Extraction complete - mutex flag stays true to prevent re-extraction
+              // The deviceCodeExtracted flag will prevent future attempts
+              extractionInProgress = false;
+            } else {
+              // No device code found yet, allow next data chunk to try again
+              extractionInProgress = false;
+            }
+          };
 
           ghProcess.stdout?.on('data', (data) => {
             const chunk = data.toString();
             output += chunk;
             debugLog('gh stdout:', chunk);
+            // Try to extract device code as data comes in
+            // Use void to explicitly ignore promise
+            void tryExtractAndOpenBrowser();
           });
 
           ghProcess.stderr?.on('data', (data) => {
             const chunk = data.toString();
             errorOutput += chunk;
             debugLog('gh stderr:', chunk);
+            // gh often outputs to stderr, so check there too
+            void tryExtractAndOpenBrowser();
           });
 
           ghProcess.on('close', (code) => {
@@ -154,17 +283,39 @@ export function registerStartGhAuth(): void {
             debugLog('Full stderr:', errorOutput);
 
             if (code === 0) {
+              // Success case - include fallbackUrl if browser failed to open
+              // so the user can manually navigate if needed
               resolve({
                 success: true,
                 data: {
                   success: true,
-                  message: 'Successfully authenticated with GitHub'
+                  message: browserOpenedSuccessfully
+                    ? 'Successfully authenticated with GitHub'
+                    : 'Authentication successful. Browser could not be opened automatically.',
+                  deviceCode: extractedDeviceCode || undefined,
+                  authUrl: extractedAuthUrl,
+                  browserOpened: browserOpenedSuccessfully,
+                  // Provide fallback URL when browser failed to open
+                  fallbackUrl: !browserOpenedSuccessfully ? extractedAuthUrl : undefined
                 }
               });
             } else {
+              // Even if auth failed, return device code info if we extracted it
+              // This allows user to retry manually with the fallback URL
+              const fallbackUrlForManualAuth = extractedDeviceCode ? extractedAuthUrl : GITHUB_DEVICE_URL;
+
               resolve({
                 success: false,
-                error: errorOutput || `Authentication failed with exit code ${code}`
+                error: errorOutput || `Authentication failed with exit code ${code}`,
+                data: {
+                  success: false,
+                  deviceCode: extractedDeviceCode || undefined,
+                  authUrl: extractedAuthUrl,
+                  browserOpened: browserOpenedSuccessfully,
+                  // Always provide fallback URL on failure for manual recovery
+                  fallbackUrl: fallbackUrlForManualAuth,
+                  message: 'Authentication failed. Please visit the URL manually to complete authentication.'
+                }
               });
             }
           });
@@ -173,14 +324,28 @@ export function registerStartGhAuth(): void {
             debugLog('gh process error:', error.message);
             resolve({
               success: false,
-              error: error.message
+              error: error.message,
+              data: {
+                success: false,
+                browserOpened: false,
+                // Provide fallback URL so user can attempt manual auth
+                fallbackUrl: GITHUB_DEVICE_URL,
+                message: 'Failed to start GitHub CLI. Please visit the URL manually to authenticate.'
+              }
             });
           });
         } catch (error) {
           debugLog('Exception in startGitHubAuth:', error instanceof Error ? error.message : error);
           resolve({
             success: false,
-            error: error instanceof Error ? error.message : 'Unknown error'
+            error: error instanceof Error ? error.message : 'Unknown error',
+            data: {
+              success: false,
+              browserOpened: false,
+              // Provide fallback URL for manual authentication recovery
+              fallbackUrl: GITHUB_DEVICE_URL,
+              message: 'An unexpected error occurred. Please visit the URL manually to authenticate.'
+            }
           });
         }
       });
@@ -366,7 +531,7 @@ export function registerGetGitHubBranches(): void {
     IPC_CHANNELS.GITHUB_GET_BRANCHES,
     async (_event: Electron.IpcMainInvokeEvent, repo: string, _token: string): Promise<IPCResult<string[]>> => {
       debugLog('getGitHubBranches handler called', { repo });
-      
+
       // Validate repo format to prevent command injection
       if (!isValidGitHubRepo(repo)) {
         debugLog('Invalid repo format rejected:', repo);
@@ -375,7 +540,7 @@ export function registerGetGitHubBranches(): void {
           error: 'Invalid repository format. Expected: owner/repo'
         };
       }
-      
+
       try {
         // Use gh CLI to list branches (uses authenticated session)
         // Use execFileSync with separate arguments to avoid shell injection
@@ -408,6 +573,203 @@ export function registerGetGitHubBranches(): void {
   );
 }
 
+/**
+ * Create a new GitHub repository using gh CLI
+ */
+export function registerCreateGitHubRepo(): void {
+  ipcMain.handle(
+    IPC_CHANNELS.GITHUB_CREATE_REPO,
+    async (
+      _event: Electron.IpcMainInvokeEvent,
+      repoName: string,
+      options: { description?: string; isPrivate?: boolean; projectPath: string; owner?: string }
+    ): Promise<IPCResult<{ fullName: string; url: string }>> => {
+      debugLog('createGitHubRepo handler called', { repoName, options });
+
+      // Validate repo name - only alphanumeric, hyphens, underscores
+      if (!/^[A-Za-z0-9_.-]+$/.test(repoName)) {
+        return {
+          success: false,
+          error: 'Invalid repository name. Use only letters, numbers, hyphens, underscores, and periods.'
+        };
+      }
+
+      try {
+        // Get the authenticated username
+        const username = execSync('gh api user --jq .login', {
+          encoding: 'utf-8',
+          stdio: 'pipe'
+        }).trim();
+
+        // Determine the owner (personal account or organization)
+        const owner = options.owner || username;
+        const isOrgRepo = owner !== username;
+
+        // Build the full repo name (owner/repo format for orgs)
+        const repoFullName = isOrgRepo ? `${owner}/${repoName}` : repoName;
+
+        // Build gh repo create command arguments
+        const args = ['repo', 'create', repoFullName, '--source', options.projectPath];
+
+        if (options.isPrivate) {
+          args.push('--private');
+        } else {
+          args.push('--public');
+        }
+
+        if (options.description) {
+          args.push('--description', options.description);
+        }
+
+        // Push to remote after creation
+        args.push('--push');
+
+        debugLog('Running: gh', args);
+        const output = execFileSync('gh', args, {
+          encoding: 'utf-8',
+          cwd: options.projectPath,
+          stdio: 'pipe'
+        });
+
+        debugLog('gh repo create output:', output);
+
+        const fullName = `${owner}/${repoName}`;
+        const url = `https://github.com/${fullName}`;
+
+        debugLog('Created repo:', { fullName, url });
+
+        return {
+          success: true,
+          data: { fullName, url }
+        };
+      } catch (error) {
+        const errorMessage = error instanceof Error ? error.message : 'Failed to create repository';
+        debugLog('Failed to create repo:', errorMessage);
+        return {
+          success: false,
+          error: errorMessage
+        };
+      }
+    }
+  );
+}
+
+/**
+ * Add a remote origin to a local git repository
+ */
+export function registerAddGitRemote(): void {
+  ipcMain.handle(
+    IPC_CHANNELS.GITHUB_ADD_REMOTE,
+    async (
+      _event: Electron.IpcMainInvokeEvent,
+      projectPath: string,
+      repoFullName: string
+    ): Promise<IPCResult<{ remoteUrl: string }>> => {
+      debugLog('addGitRemote handler called', { projectPath, repoFullName });
+
+      // Validate repo format
+      if (!isValidGitHubRepo(repoFullName)) {
+        return {
+          success: false,
+          error: 'Invalid repository format. Expected: owner/repo'
+        };
+      }
+
+      const remoteUrl = `https://github.com/${repoFullName}.git`;
+
+      try {
+        // Check if origin already exists
+        try {
+          execSync('git remote get-url origin', {
+            cwd: projectPath,
+            encoding: 'utf-8',
+            stdio: 'pipe'
+          });
+          // Origin exists, remove it first
+          debugLog('Removing existing origin remote');
+          execSync('git remote remove origin', {
+            cwd: projectPath,
+            encoding: 'utf-8',
+            stdio: 'pipe'
+          });
+        } catch {
+          // No origin exists, which is fine
+        }
+
+        // Add the remote
+        debugLog('Adding remote origin:', remoteUrl);
+        execFileSync('git', ['remote', 'add', 'origin', remoteUrl], {
+          cwd: projectPath,
+          encoding: 'utf-8',
+          stdio: 'pipe'
+        });
+
+        debugLog('Remote added successfully');
+        return {
+          success: true,
+          data: { remoteUrl }
+        };
+      } catch (error) {
+        const errorMessage = error instanceof Error ? error.message : 'Failed to add remote';
+        debugLog('Failed to add remote:', errorMessage);
+        return {
+          success: false,
+          error: errorMessage
+        };
+      }
+    }
+  );
+}
+
+/**
+ * List user's GitHub organizations
+ */
+export function registerListGitHubOrgs(): void {
+  ipcMain.handle(
+    IPC_CHANNELS.GITHUB_LIST_ORGS,
+    async (): Promise<IPCResult<{ orgs: Array<{ login: string; avatarUrl?: string }> }>> => {
+      debugLog('listGitHubOrgs handler called');
+
+      try {
+        // Get user's organizations
+        const output = execSync('gh api user/orgs --jq \'.[] | {login: .login, avatarUrl: .avatar_url}\'', {
+          encoding: 'utf-8',
+          stdio: 'pipe'
+        });
+
+        // Parse the JSON lines output
+        const orgs: Array<{ login: string; avatarUrl?: string }> = [];
+        const lines = output.trim().split('\n').filter(line => line.trim());
+
+        for (const line of lines) {
+          try {
+            const org = JSON.parse(line);
+            orgs.push({
+              login: org.login,
+              avatarUrl: org.avatarUrl
+            });
+          } catch {
+            // Skip invalid JSON lines
+          }
+        }
+
+        debugLog('Found organizations:', orgs.length);
+        return {
+          success: true,
+          data: { orgs }
+        };
+      } catch (error) {
+        const errorMessage = error instanceof Error ? error.message : 'Failed to list organizations';
+        debugLog('Failed to list orgs:', errorMessage);
+        return {
+          success: true, // Return success with empty array - user might not have any orgs
+          data: { orgs: [] }
+        };
+      }
+    }
+  );
+}
+
 /**
  * Register all GitHub OAuth handlers
  */
@@ -421,5 +783,8 @@ export function registerGithubOAuthHandlers(): void {
   registerListUserRepos();
   registerDetectGitHubRepo();
   registerGetGitHubBranches();
+  registerCreateGitHubRepo();
+  registerAddGitRemote();
+  registerListGitHubOrgs();
   debugLog('GitHub OAuth handlers registered');
 }

auto-claude-ui/src/main/ipc-handlers/github/repository-handlers.ts

@@ -6,7 +6,7 @@ import { ipcMain } from 'electron';
 import { IPC_CHANNELS } from '../../../shared/constants';
 import type { IPCResult, GitHubRepository, GitHubSyncStatus } from '../../../shared/types';
 import { projectStore } from '../../project-store';
-import { getGitHubConfig, githubFetch } from './utils';
+import { getGitHubConfig, githubFetch, normalizeRepoReference } from './utils';
 import type { GitHubAPIRepository } from './types';
 
 /**
@@ -33,16 +33,28 @@ export function registerCheckConnection(): void {
       }
 
       try {
+        // Normalize repo reference (handles full URLs, git URLs, etc.)
+        const normalizedRepo = normalizeRepoReference(config.repo);
+        if (!normalizedRepo) {
+          return {
+            success: true,
+            data: {
+              connected: false,
+              error: 'Invalid repository format. Use owner/repo or GitHub URL.'
+            }
+          };
+        }
+
         // Fetch repo info
         const repoData = await githubFetch(
           config.token,
-          `/repos/${config.repo}`
+          `/repos/${normalizedRepo}`
         ) as { full_name: string; description?: string };
 
         // Count open issues
         const issuesData = await githubFetch(
           config.token,
-          `/repos/${config.repo}/issues?state=open&per_page=1`
+          `/repos/${normalizedRepo}/issues?state=open&per_page=1`
         ) as unknown[];
 
         const openCount = Array.isArray(issuesData) ? issuesData.length : 0;
@@ -71,7 +83,7 @@ export function registerCheckConnection(): void {
 }
 
 /**
- * Get list of GitHub repositories
+ * Get list of GitHub repositories (personal + organization)
  */
 export function registerGetRepositories(): void {
   ipcMain.handle(
@@ -88,9 +100,11 @@ export function registerGetRepositories(): void {
       }
 
       try {
+        // Fetch user's personal + organization repos
+        // affiliation parameter includes: owner, collaborator, organization_member
         const repos = await githubFetch(
           config.token,
-          '/user/repos?per_page=100&sort=updated'
+          '/user/repos?per_page=100&sort=updated&affiliation=owner,collaborator,organization_member'
         ) as GitHubAPIRepository[];
 
         const result: GitHubRepository[] = repos.map(repo => ({

auto-claude-ui/src/main/ipc-handlers/github/utils.ts

@@ -54,6 +54,32 @@ export function getGitHubConfig(project: Project): GitHubConfig | null {
   }
 }
 
+/**
+ * Normalize a GitHub repository reference to owner/repo format
+ * Handles:
+ * - owner/repo (already normalized)
+ * - https://github.com/owner/repo
+ * - https://github.com/owner/repo.git
+ * - git@github.com:owner/repo.git
+ */
+export function normalizeRepoReference(repo: string): string {
+  if (!repo) return '';
+
+  // Remove trailing .git if present
+  let normalized = repo.replace(/\.git$/, '');
+
+  // Handle full GitHub URLs
+  if (normalized.startsWith('https://github.com/')) {
+    normalized = normalized.replace('https://github.com/', '');
+  } else if (normalized.startsWith('http://github.com/')) {
+    normalized = normalized.replace('http://github.com/', '');
+  } else if (normalized.startsWith('git@github.com:')) {
+    normalized = normalized.replace('git@github.com:', '');
+  }
+
+  return normalized.trim();
+}
+
 /**
  * Make a request to the GitHub API
  */

auto-claude-ui/src/main/ipc-handlers/ideation/generation-handlers.ts

@@ -3,11 +3,45 @@
  */
 
 import type { IpcMainEvent, IpcMainInvokeEvent, BrowserWindow } from 'electron';
-import { IPC_CHANNELS } from '../../../shared/constants';
-import type { IPCResult, IdeationConfig, IdeationGenerationStatus } from '../../../shared/types';
+import { app } from 'electron';
+import { existsSync, readFileSync } from 'fs';
+import path from 'path';
+import { IPC_CHANNELS, DEFAULT_APP_SETTINGS, DEFAULT_FEATURE_MODELS, DEFAULT_FEATURE_THINKING } from '../../../shared/constants';
+import type { IPCResult, IdeationConfig, IdeationGenerationStatus, AppSettings } from '../../../shared/types';
 import { projectStore } from '../../project-store';
 import type { AgentManager } from '../../agent';
-import { debugLog } from '../../../shared/utils/debug-logger';
+import { debugLog, debugError } from '../../../shared/utils/debug-logger';
+
+/**
+ * Read ideation feature settings from the settings file
+ */
+function getIdeationFeatureSettings(): { model?: string; thinkingLevel?: string } {
+  const settingsPath = path.join(app.getPath('userData'), 'settings.json');
+
+  try {
+    if (existsSync(settingsPath)) {
+      const content = readFileSync(settingsPath, 'utf-8');
+      const settings: AppSettings = { ...DEFAULT_APP_SETTINGS, ...JSON.parse(content) };
+
+      // Get ideation-specific settings
+      const featureModels = settings.featureModels || DEFAULT_FEATURE_MODELS;
+      const featureThinking = settings.featureThinking || DEFAULT_FEATURE_THINKING;
+
+      return {
+        model: featureModels.ideation,
+        thinkingLevel: featureThinking.ideation
+      };
+    }
+  } catch (error) {
+    debugError('[Ideation Handler] Failed to read feature settings:', error);
+  }
+
+  // Return defaults if settings file doesn't exist or fails to parse
+  return {
+    model: DEFAULT_FEATURE_MODELS.ideation,
+    thinkingLevel: DEFAULT_FEATURE_THINKING.ideation
+  };
+}
 
 /**
  * Start ideation generation for a project
@@ -19,10 +53,20 @@ export function startIdeationGeneration(
   agentManager: AgentManager,
   mainWindow: BrowserWindow | null
 ): void {
+  // Get feature settings and merge with config
+  const featureSettings = getIdeationFeatureSettings();
+  const configWithSettings: IdeationConfig = {
+    ...config,
+    model: config.model || featureSettings.model,
+    thinkingLevel: config.thinkingLevel || featureSettings.thinkingLevel
+  };
+
   debugLog('[Ideation Handler] Start generation request:', {
     projectId,
-    enabledTypes: config.enabledTypes,
-    maxIdeasPerType: config.maxIdeasPerType
+    enabledTypes: configWithSettings.enabledTypes,
+    maxIdeasPerType: configWithSettings.maxIdeasPerType,
+    model: configWithSettings.model,
+    thinkingLevel: configWithSettings.thinkingLevel
   });
 
   if (!mainWindow) return;
@@ -40,11 +84,13 @@ export function startIdeationGeneration(
 
   debugLog('[Ideation Handler] Starting agent manager generation:', {
     projectId,
-    projectPath: project.path
+    projectPath: project.path,
+    model: configWithSettings.model,
+    thinkingLevel: configWithSettings.thinkingLevel
   });
 
   // Start ideation generation via agent manager
-  agentManager.startIdeationGeneration(projectId, project.path, config, false);
+  agentManager.startIdeationGeneration(projectId, project.path, configWithSettings, false);
 
   // Send initial progress
   mainWindow.webContents.send(
@@ -68,6 +114,20 @@ export function refreshIdeationSession(
   agentManager: AgentManager,
   mainWindow: BrowserWindow | null
 ): void {
+  // Get feature settings and merge with config
+  const featureSettings = getIdeationFeatureSettings();
+  const configWithSettings: IdeationConfig = {
+    ...config,
+    model: config.model || featureSettings.model,
+    thinkingLevel: config.thinkingLevel || featureSettings.thinkingLevel
+  };
+
+  debugLog('[Ideation Handler] Refresh session request:', {
+    projectId,
+    model: configWithSettings.model,
+    thinkingLevel: configWithSettings.thinkingLevel
+  });
+
   if (!mainWindow) return;
 
   const project = projectStore.getProject(projectId);
@@ -81,7 +141,7 @@ export function refreshIdeationSession(
   }
 
   // Start ideation regeneration with refresh flag
-  agentManager.startIdeationGeneration(projectId, project.path, config, true);
+  agentManager.startIdeationGeneration(projectId, project.path, configWithSettings, true);
 
   // Send initial progress
   mainWindow.webContents.send(

auto-claude-ui/src/main/ipc-handlers/index.ts

@@ -92,7 +92,7 @@ export function setupIpcHandlers(
   // Insights handlers
   registerInsightsHandlers(getMainWindow);
 
-  // Docker & infrastructure handlers (for Graphiti/FalkorDB)
+  // Memory & infrastructure handlers (for Graphiti/LadybugDB)
   registerDockerHandlers();
 
   // App auto-update handlers

auto-claude-ui/src/main/ipc-handlers/memory-handlers.ts

@@ -0,0 +1,532 @@
+/**
+ * Memory Infrastructure IPC Handlers
+ *
+ * Provides memory database status and validation for the Graphiti integration.
+ * Uses LadybugDB (embedded Kuzu-based database) - no Docker required.
+ */
+
+import { ipcMain } from 'electron';
+import { spawn } from 'child_process';
+import * as path from 'path';
+import * as fs from 'fs';
+import { IPC_CHANNELS } from '../../shared/constants';
+import type {
+  IPCResult,
+  InfrastructureStatus,
+  GraphitiValidationResult,
+  GraphitiConnectionTestResult,
+} from '../../shared/types';
+import {
+  getMemoryServiceStatus,
+  getMemoryService,
+  getDefaultDbPath,
+  isKuzuAvailable,
+} from '../memory-service';
+import { validateOpenAIApiKey } from '../api-validation-service';
+import { findPythonCommand, parsePythonCommand } from '../python-detector';
+
+// Ollama types
+interface OllamaStatus {
+  running: boolean;
+  url: string;
+  version?: string;
+  message?: string;
+}
+
+interface OllamaModel {
+  name: string;
+  size_bytes: number;
+  size_gb: number;
+  modified_at: string;
+  is_embedding: boolean;
+  embedding_dim?: number | null;
+  description?: string;
+}
+
+interface OllamaEmbeddingModel {
+  name: string;
+  embedding_dim: number | null;
+  description: string;
+  size_bytes: number;
+  size_gb: number;
+}
+
+interface OllamaRecommendedModel {
+  name: string;
+  description: string;
+  size_estimate: string;
+  dim: number;
+  installed: boolean;
+}
+
+interface OllamaPullResult {
+  model: string;
+  status: 'completed' | 'failed';
+  output: string[];
+}
+
+/**
+ * Execute the ollama_model_detector.py script
+ */
+async function executeOllamaDetector(
+  command: string,
+  baseUrl?: string
+): Promise<{ success: boolean; data?: unknown; error?: string }> {
+  const pythonCmd = findPythonCommand();
+  if (!pythonCmd) {
+    return { success: false, error: 'Python not found' };
+  }
+
+  // Find the ollama_model_detector.py script
+  const possiblePaths = [
+    path.resolve(__dirname, '..', '..', '..', 'auto-claude', 'ollama_model_detector.py'),
+    path.resolve(process.cwd(), 'auto-claude', 'ollama_model_detector.py'),
+    path.resolve(process.cwd(), '..', 'auto-claude', 'ollama_model_detector.py'),
+  ];
+
+  let scriptPath: string | null = null;
+  for (const p of possiblePaths) {
+    if (fs.existsSync(p)) {
+      scriptPath = p;
+      break;
+    }
+  }
+
+  if (!scriptPath) {
+    return { success: false, error: 'ollama_model_detector.py script not found' };
+  }
+
+  const [pythonExe, baseArgs] = parsePythonCommand(pythonCmd);
+  const args = [...baseArgs, scriptPath, command];
+  if (baseUrl) {
+    args.push('--base-url', baseUrl);
+  }
+
+  return new Promise((resolve) => {
+    let resolved = false;
+    const proc = spawn(pythonExe, args, {
+      stdio: ['ignore', 'pipe', 'pipe'],
+    });
+
+    let stdout = '';
+    let stderr = '';
+
+    proc.stdout.on('data', (data) => {
+      stdout += data.toString();
+    });
+
+    proc.stderr.on('data', (data) => {
+      stderr += data.toString();
+    });
+
+    // Single timeout mechanism to avoid race condition
+    const timeoutId = setTimeout(() => {
+      if (!resolved) {
+        resolved = true;
+        proc.kill();
+        resolve({ success: false, error: 'Timeout' });
+      }
+    }, 10000);
+
+    proc.on('close', (code) => {
+      if (resolved) return;
+      resolved = true;
+      clearTimeout(timeoutId);
+      if (code === 0 && stdout) {
+        try {
+          resolve(JSON.parse(stdout));
+        } catch {
+          resolve({ success: false, error: `Invalid JSON: ${stdout}` });
+        }
+      } else {
+        resolve({ success: false, error: stderr || `Exit code ${code}` });
+      }
+    });
+
+    proc.on('error', (err) => {
+      if (resolved) return;
+      resolved = true;
+      clearTimeout(timeoutId);
+      resolve({ success: false, error: err.message });
+    });
+  });
+}
+
+/**
+ * Register all memory-related IPC handlers
+ */
+export function registerMemoryHandlers(): void {
+  // Get memory infrastructure status
+  ipcMain.handle(
+    IPC_CHANNELS.MEMORY_STATUS,
+    async (_): Promise<IPCResult<InfrastructureStatus>> => {
+      try {
+        const status = getMemoryServiceStatus();
+        return {
+          success: true,
+          data: {
+            memory: status,
+            ready: status.kuzuInstalled && status.databaseExists,
+          },
+        };
+      } catch (error) {
+        return {
+          success: false,
+          error: error instanceof Error ? error.message : 'Failed to check memory status',
+        };
+      }
+    }
+  );
+
+  // List available databases
+  ipcMain.handle(
+    IPC_CHANNELS.MEMORY_LIST_DATABASES,
+    async (_, dbPath?: string): Promise<IPCResult<string[]>> => {
+      try {
+        const status = getMemoryServiceStatus(dbPath);
+        return { success: true, data: status.databases };
+      } catch (error) {
+        return {
+          success: false,
+          error: error instanceof Error ? error.message : 'Failed to list databases',
+        };
+      }
+    }
+  );
+
+  // Test memory database connection
+  ipcMain.handle(
+    IPC_CHANNELS.MEMORY_TEST_CONNECTION,
+    async (_, dbPath?: string, database?: string): Promise<IPCResult<GraphitiValidationResult>> => {
+      try {
+        if (!isKuzuAvailable()) {
+          return {
+            success: true,
+            data: {
+              success: false,
+              message: 'kuzu-node is not installed. Memory features require Python 3.12+ with LadybugDB.',
+            },
+          };
+        }
+
+        const service = getMemoryService({
+          dbPath: dbPath || getDefaultDbPath(),
+          database: database || 'auto_claude_memory',
+        });
+
+        const result = await service.testConnection();
+        return { success: true, data: result };
+      } catch (error) {
+        return {
+          success: false,
+          error: error instanceof Error ? error.message : 'Failed to test connection',
+        };
+      }
+    }
+  );
+
+  // ============================================
+  // Graphiti Validation Handlers
+  // ============================================
+
+  // Validate LLM provider API key (OpenAI, Anthropic, etc.)
+  ipcMain.handle(
+    IPC_CHANNELS.GRAPHITI_VALIDATE_LLM,
+    async (_, provider: string, apiKey: string): Promise<IPCResult<GraphitiValidationResult>> => {
+      try {
+        // For now, we only validate OpenAI - other providers can be added later
+        if (provider === 'openai') {
+          const result = await validateOpenAIApiKey(apiKey);
+          return { success: true, data: result };
+        }
+
+        // For other providers, do basic validation
+        if (!apiKey || !apiKey.trim()) {
+          return {
+            success: true,
+            data: {
+              success: false,
+              message: 'API key is required',
+            },
+          };
+        }
+
+        return {
+          success: true,
+          data: {
+            success: true,
+            message: `${provider} API key format appears valid`,
+            details: { provider },
+          },
+        };
+      } catch (error) {
+        return {
+          success: false,
+          error: error instanceof Error ? error.message : 'Failed to validate API key',
+        };
+      }
+    }
+  );
+
+  // Test full Graphiti connection (Database + LLM provider)
+  ipcMain.handle(
+    IPC_CHANNELS.GRAPHITI_TEST_CONNECTION,
+    async (
+      _,
+      config: {
+        dbPath?: string;
+        database?: string;
+        llmProvider: string;
+        apiKey: string;
+      }
+    ): Promise<IPCResult<GraphitiConnectionTestResult>> => {
+      try {
+        // Test database connection
+        let databaseResult: GraphitiValidationResult;
+
+        if (!isKuzuAvailable()) {
+          databaseResult = {
+            success: false,
+            message: 'kuzu-node is not installed. Memory features require Python 3.12+ with LadybugDB.',
+          };
+        } else {
+          const service = getMemoryService({
+            dbPath: config.dbPath || getDefaultDbPath(),
+            database: config.database || 'auto_claude_memory',
+          });
+          databaseResult = await service.testConnection();
+        }
+
+        // Test LLM provider
+        let llmResult: GraphitiValidationResult;
+
+        if (config.llmProvider === 'openai') {
+          llmResult = await validateOpenAIApiKey(config.apiKey);
+        } else if (config.llmProvider === 'ollama') {
+          // Ollama doesn't need API key validation
+          llmResult = {
+            success: true,
+            message: 'Ollama (local) does not require API key validation',
+            details: { provider: 'ollama' },
+          };
+        } else {
+          // Basic validation for other providers
+          llmResult = config.apiKey && config.apiKey.trim()
+            ? {
+                success: true,
+                message: `${config.llmProvider} API key format appears valid`,
+                details: { provider: config.llmProvider },
+              }
+            : {
+                success: false,
+                message: 'API key is required',
+              };
+        }
+
+        return {
+          success: true,
+          data: {
+            database: databaseResult,
+            llmProvider: llmResult,
+            ready: databaseResult.success && llmResult.success,
+          },
+        };
+      } catch (error) {
+        return {
+          success: false,
+          error: error instanceof Error ? error.message : 'Failed to test Graphiti connection',
+        };
+      }
+    }
+  );
+
+  // ============================================
+  // Ollama Model Detection Handlers
+  // ============================================
+
+  // Check if Ollama is running
+  ipcMain.handle(
+    IPC_CHANNELS.OLLAMA_CHECK_STATUS,
+    async (_, baseUrl?: string): Promise<IPCResult<OllamaStatus>> => {
+      try {
+        const result = await executeOllamaDetector('check-status', baseUrl);
+
+        if (!result.success) {
+          return {
+            success: false,
+            error: result.error || 'Failed to check Ollama status',
+          };
+        }
+
+        return {
+          success: true,
+          data: result.data as OllamaStatus,
+        };
+      } catch (error) {
+        return {
+          success: false,
+          error: error instanceof Error ? error.message : 'Failed to check Ollama status',
+        };
+      }
+    }
+  );
+
+  // List all Ollama models
+  ipcMain.handle(
+    IPC_CHANNELS.OLLAMA_LIST_MODELS,
+    async (_, baseUrl?: string): Promise<IPCResult<{ models: OllamaModel[]; count: number }>> => {
+      try {
+        const result = await executeOllamaDetector('list-models', baseUrl);
+
+        if (!result.success) {
+          return {
+            success: false,
+            error: result.error || 'Failed to list Ollama models',
+          };
+        }
+
+        const data = result.data as { models: OllamaModel[]; count: number; url: string };
+        return {
+          success: true,
+          data: {
+            models: data.models,
+            count: data.count,
+          },
+        };
+      } catch (error) {
+        return {
+          success: false,
+          error: error instanceof Error ? error.message : 'Failed to list Ollama models',
+        };
+      }
+    }
+  );
+
+  // List only embedding models from Ollama
+  ipcMain.handle(
+    IPC_CHANNELS.OLLAMA_LIST_EMBEDDING_MODELS,
+    async (
+      _,
+      baseUrl?: string
+    ): Promise<IPCResult<{ embedding_models: OllamaEmbeddingModel[]; count: number }>> => {
+      try {
+        const result = await executeOllamaDetector('list-embedding-models', baseUrl);
+
+        if (!result.success) {
+          return {
+            success: false,
+            error: result.error || 'Failed to list Ollama embedding models',
+          };
+        }
+
+        const data = result.data as {
+          embedding_models: OllamaEmbeddingModel[];
+          count: number;
+          url: string;
+        };
+        return {
+          success: true,
+          data: {
+            embedding_models: data.embedding_models,
+            count: data.count,
+          },
+        };
+      } catch (error) {
+        return {
+          success: false,
+          error: error instanceof Error ? error.message : 'Failed to list embedding models',
+        };
+      }
+    }
+  );
+
+  // Pull (download) an Ollama model
+  ipcMain.handle(
+    IPC_CHANNELS.OLLAMA_PULL_MODEL,
+    async (
+      _,
+      modelName: string,
+      baseUrl?: string
+    ): Promise<IPCResult<OllamaPullResult>> => {
+      try {
+        const pythonCmd = findPythonCommand();
+        if (!pythonCmd) {
+          return { success: false, error: 'Python not found' };
+        }
+
+        // Find the ollama_model_detector.py script
+        const possiblePaths = [
+          path.resolve(__dirname, '..', '..', '..', 'auto-claude', 'ollama_model_detector.py'),
+          path.resolve(process.cwd(), 'auto-claude', 'ollama_model_detector.py'),
+          path.resolve(process.cwd(), '..', 'auto-claude', 'ollama_model_detector.py'),
+        ];
+
+        let scriptPath: string | null = null;
+        for (const p of possiblePaths) {
+          if (fs.existsSync(p)) {
+            scriptPath = p;
+            break;
+          }
+        }
+
+        if (!scriptPath) {
+          return { success: false, error: 'ollama_model_detector.py script not found' };
+        }
+
+        const [pythonExe, baseArgs] = parsePythonCommand(pythonCmd);
+        const args = [...baseArgs, scriptPath, 'pull-model', modelName];
+
+        return new Promise((resolve) => {
+          const proc = spawn(pythonExe, args, {
+            stdio: ['ignore', 'pipe', 'pipe'],
+            timeout: 600000, // 10 minute timeout for large models
+          });
+
+          let stdout = '';
+          let stderr = '';
+
+          proc.stdout.on('data', (data) => {
+            stdout += data.toString();
+          });
+
+          proc.stderr.on('data', (data) => {
+            stderr += data.toString();
+            // Could emit progress events here in the future
+          });
+
+          proc.on('close', (code) => {
+            if (code === 0 && stdout) {
+              try {
+                const result = JSON.parse(stdout);
+                if (result.success) {
+                  resolve({
+                    success: true,
+                    data: result.data as OllamaPullResult,
+                  });
+                } else {
+                  resolve({
+                    success: false,
+                    error: result.error || 'Failed to pull model',
+                  });
+                }
+              } catch {
+                resolve({ success: false, error: `Invalid JSON: ${stdout}` });
+              }
+            } else {
+              resolve({ success: false, error: stderr || `Exit code ${code}` });
+            }
+          });
+
+          proc.on('error', (err) => {
+            resolve({ success: false, error: err.message });
+          });
+        });
+      } catch (error) {
+        return {
+          success: false,
+          error: error instanceof Error ? error.message : 'Failed to pull model',
+        };
+      }
+    }
+  );
+}

auto-claude-ui/src/main/ipc-handlers/project-handlers.ts

@@ -26,6 +26,7 @@ import { changelogService } from '../changelog-service';
 import { insightsService } from '../insights-service';
 import { titleGenerator } from '../title-generator';
 import type { BrowserWindow } from 'electron';
+import { getEffectiveSourcePath } from '../updater/path-resolver';
 
 // ============================================
 // Git Helper Functions
@@ -102,93 +103,6 @@ function detectMainBranch(projectPath: string): string | null {
 
 const settingsPath = path.join(app.getPath('userData'), 'settings.json');
 
-/**
- * Auto-detect the auto-claude source path relative to the app location.
- * Works across platforms (macOS, Windows, Linux) in both dev and production modes.
- */
-const detectAutoBuildSourcePath = (): string | null => {
-  const possiblePaths: string[] = [];
-
-  // Development mode paths
-  if (is.dev) {
-    // In dev, __dirname is typically auto-claude-ui/out/main
-    // We need to go up to the project root to find auto-claude/
-    possiblePaths.push(
-      path.resolve(__dirname, '..', '..', '..', 'auto-claude'),  // From out/main up 3 levels
-      path.resolve(__dirname, '..', '..', 'auto-claude'),        // From out/main up 2 levels
-      path.resolve(process.cwd(), 'auto-claude'),                // From cwd (project root)
-      path.resolve(process.cwd(), '..', 'auto-claude')           // From cwd parent (if running from auto-claude-ui/)
-    );
-  } else {
-    // Production mode paths (packaged app)
-    // On Windows/Linux/macOS, the app might be installed anywhere
-    // We check common locations relative to the app bundle
-    const appPath = app.getAppPath();
-    possiblePaths.push(
-      path.resolve(appPath, '..', 'auto-claude'),               // Sibling to app
-      path.resolve(appPath, '..', '..', 'auto-claude'),         // Up 2 from app
-      path.resolve(appPath, '..', '..', '..', 'auto-claude'),   // Up 3 from app
-      path.resolve(process.resourcesPath, '..', 'auto-claude'), // Relative to resources
-      path.resolve(process.resourcesPath, '..', '..', 'auto-claude')
-    );
-  }
-
-  // Add process.cwd() as last resort on all platforms
-  possiblePaths.push(path.resolve(process.cwd(), 'auto-claude'));
-
-  // Enable debug logging with AUTO_CLAUDE_DEBUG=1
-  const debug = process.env.AUTO_CLAUDE_DEBUG === '1' || process.env.AUTO_CLAUDE_DEBUG === 'true';
-
-  if (debug) {
-    console.warn('[project-handlers:detectAutoBuildSourcePath] Platform:', process.platform);
-    console.warn('[project-handlers:detectAutoBuildSourcePath] Is dev:', is.dev);
-    console.warn('[project-handlers:detectAutoBuildSourcePath] __dirname:', __dirname);
-    console.warn('[project-handlers:detectAutoBuildSourcePath] app.getAppPath():', app.getAppPath());
-    console.warn('[project-handlers:detectAutoBuildSourcePath] process.cwd():', process.cwd());
-    console.warn('[project-handlers:detectAutoBuildSourcePath] Checking paths:', possiblePaths);
-  }
-
-  for (const p of possiblePaths) {
-    // Use requirements.txt as marker - it always exists in auto-claude source
-    const markerPath = path.join(p, 'requirements.txt');
-    const exists = existsSync(p) && existsSync(markerPath);
-
-    if (debug) {
-      console.warn(`[project-handlers:detectAutoBuildSourcePath] Checking ${p}: ${exists ? '✓ FOUND' : '✗ not found'}`);
-    }
-
-    if (exists) {
-      console.warn(`[project-handlers:detectAutoBuildSourcePath] Auto-detected source path: ${p}`);
-      return p;
-    }
-  }
-
-  console.warn('[project-handlers:detectAutoBuildSourcePath] Could not auto-detect Auto Claude source path.');
-  console.warn('[project-handlers:detectAutoBuildSourcePath] Set AUTO_CLAUDE_DEBUG=1 environment variable for detailed path checking.');
-  return null;
-};
-
-/**
- * Get the configured auto-claude source path from settings, or auto-detect
- */
-const getAutoBuildSourcePath = (): string | null => {
-  // First check if manually configured
-  if (existsSync(settingsPath)) {
-    try {
-      const content = readFileSync(settingsPath, 'utf-8');
-      const settings = JSON.parse(content);
-      if (settings.autoBuildPath && existsSync(settings.autoBuildPath)) {
-        return settings.autoBuildPath;
-      }
-    } catch {
-      // Fall through to auto-detect
-    }
-  }
-
-  // Auto-detect from app location
-  return detectAutoBuildSourcePath();
-};
-
 /**
  * Configure all Python-dependent services with the managed Python path
  */
@@ -211,7 +125,7 @@ const initializePythonEnvironment = async (
   pythonEnvManager: PythonEnvManager,
   agentManager: AgentManager
 ): Promise<PythonEnvStatus> => {
-  const autoBuildSource = getAutoBuildSourcePath();
+  const autoBuildSource = getEffectiveSourcePath();
   if (!autoBuildSource) {
     console.warn('[IPC] Auto-build source not found, skipping Python env init');
     return {
@@ -304,6 +218,31 @@ export function registerProjectHandlers(
     }
   );
 
+  // ============================================
+  // Tab State Operations (persisted in main process)
+  // ============================================
+
+  ipcMain.handle(
+    IPC_CHANNELS.TAB_STATE_GET,
+    async (): Promise<IPCResult<{ openProjectIds: string[]; activeProjectId: string | null; tabOrder: string[] }>> => {
+      const tabState = projectStore.getTabState();
+      console.log('[IPC] TAB_STATE_GET returning:', tabState);
+      return { success: true, data: tabState };
+    }
+  );
+
+  ipcMain.handle(
+    IPC_CHANNELS.TAB_STATE_SAVE,
+    async (
+      _,
+      tabState: { openProjectIds: string[]; activeProjectId: string | null; tabOrder: string[] }
+    ): Promise<IPCResult> => {
+      console.log('[IPC] TAB_STATE_SAVE called with:', tabState);
+      projectStore.saveTabState(tabState);
+      return { success: true };
+    }
+  );
+
   // ============================================
   // Project Initialization Operations
   // ============================================

auto-claude-ui/src/main/ipc-handlers/roadmap-handlers.ts

@@ -1,13 +1,45 @@
-import { ipcMain } from 'electron';
+import { ipcMain, app } from 'electron';
 import type { BrowserWindow } from 'electron';
-import { IPC_CHANNELS, AUTO_BUILD_PATHS, getSpecsDir } from '../../shared/constants';
-import type { IPCResult, Roadmap, RoadmapFeature, RoadmapFeatureStatus, RoadmapGenerationStatus, Task, TaskMetadata, CompetitorAnalysis } from '../../shared/types';
+import { IPC_CHANNELS, AUTO_BUILD_PATHS, getSpecsDir, DEFAULT_APP_SETTINGS, DEFAULT_FEATURE_MODELS, DEFAULT_FEATURE_THINKING } from '../../shared/constants';
+import type { IPCResult, Roadmap, RoadmapFeature, RoadmapFeatureStatus, RoadmapGenerationStatus, Task, TaskMetadata, CompetitorAnalysis, AppSettings } from '../../shared/types';
+import type { RoadmapConfig } from '../agent/types';
 import path from 'path';
 import { existsSync, readFileSync, writeFileSync, mkdirSync, readdirSync } from 'fs';
 import { projectStore } from '../project-store';
 import { AgentManager } from '../agent';
 import { debugLog, debugError } from '../../shared/utils/debug-logger';
 
+/**
+ * Read feature settings from the settings file
+ */
+function getFeatureSettings(): { model?: string; thinkingLevel?: string } {
+  const settingsPath = path.join(app.getPath('userData'), 'settings.json');
+
+  try {
+    if (existsSync(settingsPath)) {
+      const content = readFileSync(settingsPath, 'utf-8');
+      const settings: AppSettings = { ...DEFAULT_APP_SETTINGS, ...JSON.parse(content) };
+
+      // Get roadmap-specific settings
+      const featureModels = settings.featureModels || DEFAULT_FEATURE_MODELS;
+      const featureThinking = settings.featureThinking || DEFAULT_FEATURE_THINKING;
+
+      return {
+        model: featureModels.roadmap,
+        thinkingLevel: featureThinking.roadmap
+      };
+    }
+  } catch (error) {
+    debugError('[Roadmap Handler] Failed to read feature settings:', error);
+  }
+
+  // Return defaults if settings file doesn't exist or fails to parse
+  return {
+    model: DEFAULT_FEATURE_MODELS.roadmap,
+    thinkingLevel: DEFAULT_FEATURE_THINKING.roadmap
+  };
+}
+
 
 /**
  * Register all roadmap-related IPC handlers
@@ -138,7 +170,7 @@ export function registerRoadmapHandlers(
             impact: feature.impact || 'medium',
             phaseId: feature.phase_id,
             dependencies: feature.dependencies || [],
-            status: feature.status || 'idea',
+            status: feature.status || 'under_review',
             acceptanceCriteria: feature.acceptance_criteria || [],
             userStories: feature.user_stories || [],
             linkedSpecId: feature.linked_spec_id,
@@ -172,10 +204,19 @@ export function registerRoadmapHandlers(
 
   ipcMain.on(
     IPC_CHANNELS.ROADMAP_GENERATE,
-    (_, projectId: string, enableCompetitorAnalysis?: boolean) => {
+    (_, projectId: string, enableCompetitorAnalysis?: boolean, refreshCompetitorAnalysis?: boolean) => {
+      // Get feature settings for roadmap
+      const featureSettings = getFeatureSettings();
+      const config: RoadmapConfig = {
+        model: featureSettings.model,
+        thinkingLevel: featureSettings.thinkingLevel
+      };
+
       debugLog('[Roadmap Handler] Generate request:', {
         projectId,
-        enableCompetitorAnalysis
+        enableCompetitorAnalysis,
+        refreshCompetitorAnalysis,
+        config
       });
 
       const mainWindow = getMainWindow();
@@ -194,11 +235,19 @@ export function registerRoadmapHandlers(
 
       debugLog('[Roadmap Handler] Starting agent manager generation:', {
         projectId,
-        projectPath: project.path
+        projectPath: project.path,
+        config
       });
 
       // Start roadmap generation via agent manager
-      agentManager.startRoadmapGeneration(projectId, project.path, false, enableCompetitorAnalysis ?? false);
+      agentManager.startRoadmapGeneration(
+        projectId,
+        project.path,
+        false, // refresh (not a refresh operation)
+        enableCompetitorAnalysis ?? false,
+        refreshCompetitorAnalysis ?? false,
+        config
+      );
 
       // Send initial progress
       mainWindow.webContents.send(
@@ -215,7 +264,21 @@ export function registerRoadmapHandlers(
 
   ipcMain.on(
     IPC_CHANNELS.ROADMAP_REFRESH,
-    (_, projectId: string, enableCompetitorAnalysis?: boolean) => {
+    (_, projectId: string, enableCompetitorAnalysis?: boolean, refreshCompetitorAnalysis?: boolean) => {
+      // Get feature settings for roadmap
+      const featureSettings = getFeatureSettings();
+      const config: RoadmapConfig = {
+        model: featureSettings.model,
+        thinkingLevel: featureSettings.thinkingLevel
+      };
+
+      debugLog('[Roadmap Handler] Refresh request:', {
+        projectId,
+        enableCompetitorAnalysis,
+        refreshCompetitorAnalysis,
+        config
+      });
+
       const mainWindow = getMainWindow();
       if (!mainWindow) return;
 
@@ -230,7 +293,14 @@ export function registerRoadmapHandlers(
       }
 
       // Start roadmap regeneration with refresh flag
-      agentManager.startRoadmapGeneration(projectId, project.path, true, enableCompetitorAnalysis ?? false);
+      agentManager.startRoadmapGeneration(
+        projectId,
+        project.path,
+        true, // refresh (this is a refresh operation)
+        enableCompetitorAnalysis ?? false,
+        refreshCompetitorAnalysis ?? false,
+        config
+      );
 
       // Send initial progress
       mainWindow.webContents.send(
@@ -275,7 +345,7 @@ export function registerRoadmapHandlers(
     async (
       _,
       projectId: string,
-      features: RoadmapFeature[]
+      roadmapData: Roadmap
     ): Promise<IPCResult> => {
       const project = projectStore.getProject(projectId);
       if (!project) {
@@ -294,10 +364,10 @@ export function registerRoadmapHandlers(
 
       try {
         const content = readFileSync(roadmapPath, 'utf-8');
-        const roadmap = JSON.parse(content);
+        const existingRoadmap = JSON.parse(content);
 
         // Transform camelCase features back to snake_case for JSON file
-        roadmap.features = features.map((feature) => ({
+        existingRoadmap.features = roadmapData.features.map((feature) => ({
           id: feature.id,
           title: feature.title,
           description: feature.description,
@@ -315,10 +385,10 @@ export function registerRoadmapHandlers(
         }));
 
         // Update metadata timestamp
-        roadmap.metadata = roadmap.metadata || {};
-        roadmap.metadata.updated_at = new Date().toISOString();
+        existingRoadmap.metadata = existingRoadmap.metadata || {};
+        existingRoadmap.metadata.updated_at = new Date().toISOString();
 
-        writeFileSync(roadmapPath, JSON.stringify(roadmap, null, 2));
+        writeFileSync(roadmapPath, JSON.stringify(existingRoadmap, null, 2));
 
         return { success: true };
       } catch (error) {

auto-claude-ui/src/main/ipc-handlers/sections/context-roadmap-section.txt

@@ -874,4 +874,3 @@ ${(feature.acceptance_criteria || []).map((c: string) => `- [ ] ${c}`).join('\n'
       return { success: true, data: memories };
     }
   );
-

auto-claude-ui/src/main/ipc-handlers/sections/ideation-insights-section.txt

@@ -1172,4 +1172,3 @@ ${idea.rationale}
       return { success: false, error: 'Failed to rename session' };
     }
   );
-

auto-claude-ui/src/main/ipc-handlers/sections/integration-section.txt

@@ -1650,4 +1650,3 @@ ${issue.body || 'No description provided.'}
       }
     }
   );
-

auto-claude-ui/src/main/ipc-handlers/sections/task-section.txt

@@ -876,7 +876,7 @@
           try {
             // Set status to in_progress for the restart
             newStatus = 'in_progress';
-            
+
             // Update plan status for restart
             if (plan) {
               plan.status = 'in_progress';
@@ -897,7 +897,7 @@
             const specsBaseDir = getSpecsDir(project.autoBuildPath);
             const specDirForWatcher = path.join(project.path, specsBaseDir, task.specId);
             fileWatcher.watch(taskId, specDirForWatcher);
-            
+
             agentManager.startTaskExecution(
               taskId,
               project.path,
@@ -932,7 +932,7 @@
             taskId,
             recovered: true,
             newStatus,
-            message: autoRestarted 
+            message: autoRestarted
               ? 'Task recovered and restarted successfully'
               : `Task recovered successfully and moved to ${newStatus}`,
             autoRestarted
@@ -1494,4 +1494,3 @@
       }
     }
   );
-

auto-claude-ui/src/main/ipc-handlers/sections/task_extracted.txt

@@ -876,7 +876,7 @@
           try {
             // Set status to in_progress for the restart
             newStatus = 'in_progress';
-            
+
             // Update plan status for restart
             if (plan) {
               plan.status = 'in_progress';
@@ -897,7 +897,7 @@
             const specsBaseDir = getSpecsDir(project.autoBuildPath);
             const specDirForWatcher = path.join(project.path, specsBaseDir, task.specId);
             fileWatcher.watch(taskId, specDirForWatcher);
-            
+
             agentManager.startTaskExecution(
               taskId,
               project.path,
@@ -932,7 +932,7 @@
             taskId,
             recovered: true,
             newStatus,
-            message: autoRestarted 
+            message: autoRestarted
               ? 'Task recovered and restarted successfully'
               : `Task recovered successfully and moved to ${newStatus}`,
             autoRestarted

auto-claude-ui/src/main/ipc-handlers/sections/terminal-section.txt

@@ -218,12 +218,12 @@
           });
         }
 
-        return { 
-          success: true, 
-          data: { 
+        return {
+          success: true,
+          data: {
             terminalId,
             message: `A terminal has been opened to authenticate "${profile.name}". Complete the OAuth flow in your browser, then copy the token shown in the terminal.`
-          } 
+          }
         };
       } catch (error) {
         console.error('[IPC] Failed to initialize Claude profile:', error);

auto-claude-ui/src/main/ipc-handlers/sections/terminal_extracted.txt

@@ -218,12 +218,12 @@
           });
         }
 
-        return { 
-          success: true, 
-          data: { 
+        return {
+          success: true,
+          data: {
             terminalId,
             message: `A terminal has been opened to authenticate "${profile.name}". Complete the OAuth flow in your browser, then copy the token shown in the terminal.`
-          } 
+          }
         };
       } catch (error) {
         console.error('[IPC] Failed to initialize Claude profile:', error);
@@ -484,4 +484,4 @@
         };
       }
     }
-  );
+  );

auto-claude-ui/src/main/ipc-handlers/settings-handlers.ts

@@ -10,6 +10,7 @@ import type {
 } from '../../shared/types';
 import { AgentManager } from '../agent';
 import type { BrowserWindow } from 'electron';
+import { getEffectiveVersion } from '../auto-claude-updater';
 
 const settingsPath = path.join(app.getPath('userData'), 'settings.json');
 
@@ -47,8 +48,8 @@ const detectAutoBuildSourcePath = (): string | null => {
   // Add process.cwd() as last resort on all platforms
   possiblePaths.push(path.resolve(process.cwd(), 'auto-claude'));
 
-  // Enable debug logging with AUTO_CLAUDE_DEBUG=1
-  const debug = process.env.AUTO_CLAUDE_DEBUG === '1' || process.env.AUTO_CLAUDE_DEBUG === 'true';
+  // Enable debug logging with DEBUG=1
+  const debug = process.env.DEBUG === '1' || process.env.DEBUG === 'true';
 
   if (debug) {
     console.warn('[detectAutoBuildSourcePath] Platform:', process.platform);
@@ -75,7 +76,7 @@ const detectAutoBuildSourcePath = (): string | null => {
   }
 
   console.warn('[detectAutoBuildSourcePath] Could not auto-detect Auto Claude source path. Please configure manually in settings.');
-  console.warn('[detectAutoBuildSourcePath] Set AUTO_CLAUDE_DEBUG=1 environment variable for detailed path checking.');
+  console.warn('[detectAutoBuildSourcePath] Set DEBUG=1 environment variable for detailed path checking.');
   return null;
 };
 
@@ -93,7 +94,8 @@ export function registerSettingsHandlers(
   ipcMain.handle(
     IPC_CHANNELS.SETTINGS_GET,
     async (): Promise<IPCResult<AppSettings>> => {
-      let settings = { ...DEFAULT_APP_SETTINGS };
+      let settings: AppSettings = { ...DEFAULT_APP_SETTINGS };
+      let needsSave = false;
 
       if (existsSync(settingsPath)) {
         try {
@@ -104,6 +106,18 @@ export function registerSettingsHandlers(
         }
       }
 
+      // Migration: Set agent profile to 'auto' for users who haven't made a selection (one-time)
+      // This ensures new users get the optimized 'auto' profile as the default
+      // while preserving existing user preferences
+      if (!settings._migratedAgentProfileToAuto) {
+        // Only set 'auto' if user hasn't made a selection yet
+        if (!settings.selectedAgentProfile) {
+          settings.selectedAgentProfile = 'auto';
+        }
+        settings._migratedAgentProfileToAuto = true;
+        needsSave = true;
+      }
+
       // If no manual autoBuildPath is set, try to auto-detect
       if (!settings.autoBuildPath) {
         const detectedPath = detectAutoBuildSourcePath();
@@ -112,6 +126,16 @@ export function registerSettingsHandlers(
         }
       }
 
+      // Persist migration changes
+      if (needsSave) {
+        try {
+          writeFileSync(settingsPath, JSON.stringify(settings, null, 2));
+        } catch (error) {
+          console.error('[SETTINGS_GET] Failed to persist migration:', error);
+          // Continue anyway - settings will be migrated in-memory for this session
+        }
+      }
+
       return { success: true, data: settings as AppSettings };
     }
   );
@@ -264,7 +288,10 @@ export function registerSettingsHandlers(
   // ============================================
 
   ipcMain.handle(IPC_CHANNELS.APP_VERSION, async (): Promise<string> => {
-    return app.getVersion();
+    // Use effective version which accounts for source updates
+    const version = getEffectiveVersion();
+    console.log('[settings-handlers] APP_VERSION returning:', version);
+    return version;
   });
 
   // ============================================

auto-claude-ui/src/main/ipc-handlers/task-handlers.ts.backup

@@ -951,7 +951,7 @@ export function registerTaskHandlers(
           try {
             // Set status to in_progress for the restart
             newStatus = 'in_progress';
-            
+
             // Update plan status for restart
             if (plan) {
               plan.status = 'in_progress';
@@ -964,7 +964,7 @@ export function registerTaskHandlers(
             const specsBaseDir = getSpecsDir(project.autoBuildPath);
             const specDirForWatcher = path.join(project.path, specsBaseDir, task.specId);
             fileWatcher.watch(taskId, specDirForWatcher);
-            
+
             // Note: Parallel execution is handled internally by the agent
             agentManager.startTaskExecution(
               taskId,
@@ -1000,7 +1000,7 @@ export function registerTaskHandlers(
             taskId,
             recovered: true,
             newStatus,
-            message: autoRestarted 
+            message: autoRestarted
               ? 'Task recovered and restarted successfully'
               : `Task recovered successfully and moved to ${newStatus}`,
             autoRestarted

auto-claude-ui/src/main/ipc-handlers/task/execution-handlers.ts

@@ -3,6 +3,7 @@ import { IPC_CHANNELS, AUTO_BUILD_PATHS, getSpecsDir } from '../../../shared/con
 import type { IPCResult, TaskStartOptions, TaskStatus } from '../../../shared/types';
 import path from 'path';
 import { existsSync, readFileSync, writeFileSync, mkdirSync } from 'fs';
+import { spawnSync } from 'child_process';
 import { AgentManager } from '../../agent';
 import { fileWatcher } from '../../file-watcher';
 import { findTaskAndProject } from './shared';
@@ -200,6 +201,11 @@ export function registerTaskExecutionHandlers(
         task.specId
       );
 
+      // Check if worktree exists - QA needs to run in the worktree where the build happened
+      const worktreePath = path.join(project.path, '.worktrees', task.specId);
+      const worktreeSpecDir = path.join(worktreePath, specsBaseDir, task.specId);
+      const hasWorktree = existsSync(worktreePath);
+
       if (approved) {
         // Write approval to QA report
         const qaReportPath = path.join(specDir, AUTO_BUILD_PATHS.QA_REPORT);
@@ -217,15 +223,61 @@ export function registerTaskExecutionHandlers(
           );
         }
       } else {
-        // Write feedback for QA fixer
-        const fixRequestPath = path.join(specDir, 'QA_FIX_REQUEST.md');
+        // Reset and discard all changes from worktree merge in main
+        // The worktree still has all changes, so nothing is lost
+        if (hasWorktree) {
+          // Step 1: Unstage all changes
+          const resetResult = spawnSync('git', ['reset', 'HEAD'], {
+            cwd: project.path,
+            encoding: 'utf-8',
+            stdio: 'pipe'
+          });
+          if (resetResult.status === 0) {
+            console.log('[TASK_REVIEW] Unstaged changes in main');
+          }
+
+          // Step 2: Discard all working tree changes (restore to pre-merge state)
+          const checkoutResult = spawnSync('git', ['checkout', '--', '.'], {
+            cwd: project.path,
+            encoding: 'utf-8',
+            stdio: 'pipe'
+          });
+          if (checkoutResult.status === 0) {
+            console.log('[TASK_REVIEW] Discarded working tree changes in main');
+          }
+
+          // Step 3: Clean untracked files that came from the merge
+          // IMPORTANT: Exclude .auto-claude and .worktrees directories to preserve specs and worktree data
+          const cleanResult = spawnSync('git', ['clean', '-fd', '-e', '.auto-claude', '-e', '.worktrees'], {
+            cwd: project.path,
+            encoding: 'utf-8',
+            stdio: 'pipe'
+          });
+          if (cleanResult.status === 0) {
+            console.log('[TASK_REVIEW] Cleaned untracked files in main (excluding .auto-claude and .worktrees)');
+          }
+
+          console.log('[TASK_REVIEW] Main branch restored to pre-merge state');
+        }
+
+        // Write feedback for QA fixer - write to WORKTREE spec dir if it exists
+        // The QA process runs in the worktree where the build and implementation_plan.json are
+        const targetSpecDir = hasWorktree ? worktreeSpecDir : specDir;
+        const fixRequestPath = path.join(targetSpecDir, 'QA_FIX_REQUEST.md');
+
+        console.warn('[TASK_REVIEW] Writing QA fix request to:', fixRequestPath);
+        console.warn('[TASK_REVIEW] hasWorktree:', hasWorktree, 'worktreePath:', worktreePath);
+
         writeFileSync(
           fixRequestPath,
           `# QA Fix Request\n\nStatus: REJECTED\n\n## Feedback\n\n${feedback || 'No feedback provided'}\n\nCreated at: ${new Date().toISOString()}\n`
         );
 
-        // Restart QA process with dev mode
-        agentManager.startQAProcess(taskId, project.path, task.specId);
+        // Restart QA process - use worktree path if it exists, otherwise main project
+        // The QA process needs to run where the implementation_plan.json with completed subtasks is
+        const qaProjectPath = hasWorktree ? worktreePath : project.path;
+        console.warn('[TASK_REVIEW] Starting QA process with projectPath:', qaProjectPath);
+        agentManager.startQAProcess(taskId, qaProjectPath, task.specId);
 
         const mainWindow = getMainWindow();
         if (mainWindow) {

auto-claude-ui/src/main/ipc-handlers/task/worktree-handlers.ts

@@ -2,13 +2,34 @@ import { ipcMain, BrowserWindow } from 'electron';
 import { IPC_CHANNELS, AUTO_BUILD_PATHS } from '../../../shared/constants';
 import type { IPCResult, WorktreeStatus, WorktreeDiff, WorktreeDiffFile, WorktreeMergeResult, WorktreeDiscardResult, WorktreeListResult, WorktreeListItem } from '../../../shared/types';
 import path from 'path';
-import { existsSync, readdirSync, statSync } from 'fs';
-import { execSync, spawn } from 'child_process';
+import { existsSync, readdirSync, statSync, readFileSync } from 'fs';
+import { execSync, spawn, spawnSync } from 'child_process';
 import { projectStore } from '../../project-store';
 import { PythonEnvManager } from '../../python-env-manager';
 import { getEffectiveSourcePath } from '../../auto-claude-updater';
 import { getProfileEnv } from '../../rate-limit-detector';
 import { findTaskAndProject } from './shared';
+import { findPythonCommand, parsePythonCommand } from '../../python-detector';
+
+/**
+ * Read the stored base branch from task_metadata.json
+ * This is the branch the task was created from (set by user during task creation)
+ */
+function getTaskBaseBranch(specDir: string): string | undefined {
+  try {
+    const metadataPath = path.join(specDir, 'task_metadata.json');
+    if (existsSync(metadataPath)) {
+      const metadata = JSON.parse(readFileSync(metadataPath, 'utf-8'));
+      // Return baseBranch if explicitly set (not the __project_default__ marker)
+      if (metadata.baseBranch && metadata.baseBranch !== '__project_default__') {
+        return metadata.baseBranch;
+      }
+    }
+  } catch (e) {
+    console.warn('[getTaskBaseBranch] Failed to read task metadata:', e);
+  }
+  return undefined;
+}
 
 /**
  * Register worktree management handlers
@@ -48,24 +69,25 @@ export function registerWorktreeHandlers(
             encoding: 'utf-8'
           }).trim();
 
-          // Get base branch (usually main or master)
+          // Get base branch - the current branch in the main project (where changes will be merged)
+          // This matches the Python merge logic which merges into the user's current branch
           let baseBranch = 'main';
           try {
-            // Try to get the default branch
-            baseBranch = execSync('git rev-parse --abbrev-ref origin/HEAD 2>/dev/null || echo main', {
+            baseBranch = execSync('git rev-parse --abbrev-ref HEAD', {
               cwd: project.path,
               encoding: 'utf-8'
-            }).trim().replace('origin/', '');
+            }).trim();
           } catch {
             baseBranch = 'main';
           }
 
-          // Get commit count
+          // Get commit count (cross-platform - no shell syntax)
           let commitCount = 0;
           try {
-            const countOutput = execSync(`git rev-list --count ${baseBranch}..HEAD 2>/dev/null || echo 0`, {
+            const countOutput = execSync(`git rev-list --count ${baseBranch}..HEAD`, {
               cwd: worktreePath,
-              encoding: 'utf-8'
+              encoding: 'utf-8',
+              stdio: ['pipe', 'pipe', 'pipe']
             }).trim();
             commitCount = parseInt(countOutput, 10) || 0;
           } catch {
@@ -77,10 +99,12 @@ export function registerWorktreeHandlers(
           let additions = 0;
           let deletions = 0;
 
+          let diffStat = '';
           try {
-            const diffStat = execSync(`git diff --stat ${baseBranch}...HEAD 2>/dev/null || echo ""`, {
+            diffStat = execSync(`git diff --stat ${baseBranch}...HEAD`, {
               cwd: worktreePath,
-              encoding: 'utf-8'
+              encoding: 'utf-8',
+              stdio: ['pipe', 'pipe', 'pipe']
             }).trim();
 
             // Parse the summary line (e.g., "3 files changed, 50 insertions(+), 10 deletions(-)")
@@ -144,13 +168,13 @@ export function registerWorktreeHandlers(
           return { success: false, error: 'No worktree found for this task' };
         }
 
-        // Get base branch
+        // Get base branch - the current branch in the main project (where changes will be merged)
         let baseBranch = 'main';
         try {
-          baseBranch = execSync('git rev-parse --abbrev-ref origin/HEAD 2>/dev/null || echo main', {
+          baseBranch = execSync('git rev-parse --abbrev-ref HEAD', {
             cwd: project.path,
             encoding: 'utf-8'
-          }).trim().replace('origin/', '');
+          }).trim();
         } catch {
           baseBranch = 'main';
         }
@@ -158,17 +182,21 @@ export function registerWorktreeHandlers(
         // Get the diff with file stats
         const files: WorktreeDiffFile[] = [];
 
+        let numstat = '';
+        let nameStatus = '';
         try {
-          // Get numstat for additions/deletions per file
-          const numstat = execSync(`git diff --numstat ${baseBranch}...HEAD 2>/dev/null || echo ""`, {
+          // Get numstat for additions/deletions per file (cross-platform)
+          numstat = execSync(`git diff --numstat ${baseBranch}...HEAD`, {
             cwd: worktreePath,
-            encoding: 'utf-8'
+            encoding: 'utf-8',
+            stdio: ['pipe', 'pipe', 'pipe']
           }).trim();
 
-          // Get name-status for file status
-          const nameStatus = execSync(`git diff --name-status ${baseBranch}...HEAD 2>/dev/null || echo ""`, {
+          // Get name-status for file status (cross-platform)
+          nameStatus = execSync(`git diff --name-status ${baseBranch}...HEAD`, {
             cwd: worktreePath,
-            encoding: 'utf-8'
+            encoding: 'utf-8',
+            stdio: ['pipe', 'pipe', 'pipe']
           }).trim();
 
           // Parse name-status to get file statuses
@@ -272,6 +300,31 @@ export function registerWorktreeHandlers(
         const worktreePath = path.join(project.path, '.worktrees', task.specId);
         debug('Worktree path:', worktreePath, 'exists:', existsSync(worktreePath));
 
+        // Check if changes are already staged (for stage-only mode)
+        if (options?.noCommit) {
+          const stagedResult = spawnSync('git', ['diff', '--staged', '--name-only'], {
+            cwd: project.path,
+            encoding: 'utf-8'
+          });
+
+          if (stagedResult.status === 0 && stagedResult.stdout?.trim()) {
+            const stagedFiles = stagedResult.stdout.trim().split('\n');
+            debug('Changes already staged:', stagedFiles.length, 'files');
+            // Return success - changes are already staged
+            return {
+              success: true,
+              data: {
+                success: true,
+                merged: false,
+                message: `Changes already staged (${stagedFiles.length} files). Review with git diff --staged.`,
+                staged: true,
+                alreadyStaged: true,
+                projectPath: project.path
+              }
+            };
+          }
+        }
+
         // Get git status before merge
         try {
           const gitStatusBefore = execSync('git status --short', { cwd: project.path, encoding: 'utf-8' });
@@ -294,7 +347,14 @@ export function registerWorktreeHandlers(
           args.push('--no-commit');
         }
 
-        const pythonPath = pythonEnvManager.getPythonPath() || 'python3';
+        // Add --base-branch if task was created with a specific base branch
+        const taskBaseBranch = getTaskBaseBranch(specDir);
+        if (taskBaseBranch) {
+          args.push('--base-branch', taskBaseBranch);
+          debug('Using stored base branch:', taskBaseBranch);
+        }
+
+        const pythonPath = pythonEnvManager.getPythonPath() || findPythonCommand() || 'python';
         debug('Running command:', pythonPath, args.join(' '));
         debug('Working directory:', sourcePath);
 
@@ -306,11 +366,13 @@ export function registerWorktreeHandlers(
         });
 
         return new Promise((resolve) => {
-          const MERGE_TIMEOUT_MS = 120000; // 2 minutes timeout for merge operations
+          const MERGE_TIMEOUT_MS = 600000; // 10 minutes timeout for AI merge operations with many files
           let timeoutId: NodeJS.Timeout | null = null;
           let resolved = false;
 
-          const mergeProcess = spawn(pythonPath, args, {
+          // Parse Python command to handle space-separated commands like "py -3"
+          const [pythonCommand, pythonBaseArgs] = parsePythonCommand(pythonPath);
+          const mergeProcess = spawn(pythonCommand, [...pythonBaseArgs, ...args], {
             cwd: sourcePath,
             env: {
               ...process.env,
@@ -418,14 +480,17 @@ export function registerWorktreeHandlers(
                     const specBranch = `auto-claude/${task.specId}`;
                     try {
                       // Check if current branch contains all commits from spec branch
-                      const mergeBaseResult = execSync(
-                        `git merge-base --is-ancestor ${specBranch} HEAD 2>/dev/null && echo "merged" || echo "not-merged"`,
-                        { cwd: project.path, encoding: 'utf-8' }
-                      ).trim();
-                      mergeAlreadyCommitted = mergeBaseResult === 'merged';
+                      // git merge-base --is-ancestor returns exit code 0 if true, 1 if false
+                      execSync(
+                        `git merge-base --is-ancestor ${specBranch} HEAD`,
+                        { cwd: project.path, encoding: 'utf-8', stdio: ['pipe', 'pipe', 'pipe'] }
+                      );
+                      // If we reach here, the command succeeded (exit code 0) - branch is merged
+                      mergeAlreadyCommitted = true;
                       debug('Merge already committed check:', mergeAlreadyCommitted);
                     } catch {
-                      // Branch may not exist or other error - assume not merged
+                      // Exit code 1 means not merged, or branch may not exist
+                      mergeAlreadyCommitted = false;
                       debug('Could not check merge status, assuming not merged');
                     }
                   }
@@ -472,6 +537,21 @@ export function registerWorktreeHandlers(
 
               debug('Merge result. isStageOnly:', isStageOnly, 'newStatus:', newStatus, 'staged:', staged);
 
+              // Read suggested commit message if staging succeeded
+              let suggestedCommitMessage: string | undefined;
+              if (staged) {
+                const commitMsgPath = path.join(specDir, 'suggested_commit_message.txt');
+                try {
+                  if (existsSync(commitMsgPath)) {
+                    const { readFileSync } = require('fs');
+                    suggestedCommitMessage = readFileSync(commitMsgPath, 'utf-8').trim();
+                    debug('Read suggested commit message:', suggestedCommitMessage?.substring(0, 100));
+                  }
+                } catch (e) {
+                  debug('Failed to read suggested commit message:', e);
+                }
+              }
+
               // Persist the status change to implementation_plan.json
               const planPath = path.join(specDir, AUTO_BUILD_PATHS.IMPLEMENTATION_PLAN);
               try {
@@ -503,7 +583,8 @@ export function registerWorktreeHandlers(
                   success: true,
                   message,
                   staged,
-                  projectPath: staged ? project.path : undefined
+                  projectPath: staged ? project.path : undefined,
+                  suggestedCommitMessage
                 }
               });
             } else {
@@ -615,6 +696,7 @@ export function registerWorktreeHandlers(
         }
 
         const runScript = path.join(sourcePath, 'run.py');
+        const specDir = path.join(project.path, project.autoBuildPath || '.auto-claude', 'specs', task.specId);
         const args = [
           runScript,
           '--spec', task.specId,
@@ -622,14 +704,23 @@ export function registerWorktreeHandlers(
           '--merge-preview'
         ];
 
-        const pythonPath = pythonEnvManager.getPythonPath() || 'python3';
+        // Add --base-branch if task was created with a specific base branch
+        const taskBaseBranch = getTaskBaseBranch(specDir);
+        if (taskBaseBranch) {
+          args.push('--base-branch', taskBaseBranch);
+          console.warn('[IPC] Using stored base branch for preview:', taskBaseBranch);
+        }
+
+        const pythonPath = pythonEnvManager.getPythonPath() || findPythonCommand() || 'python';
         console.warn('[IPC] Running merge preview:', pythonPath, args.join(' '));
 
         // Get profile environment for consistency
         const previewProfileEnv = getProfileEnv();
 
         return new Promise((resolve) => {
-          const previewProcess = spawn(pythonPath, args, {
+          // Parse Python command to handle space-separated commands like "py -3"
+          const [pythonCommand, pythonBaseArgs] = parsePythonCommand(pythonPath);
+          const previewProcess = spawn(pythonCommand, [...pythonBaseArgs, ...args], {
             cwd: sourcePath,
             env: { ...process.env, ...previewProfileEnv, PYTHONUNBUFFERED: '1', PYTHONIOENCODING: 'utf-8', PYTHONUTF8: '1', DEBUG: 'true' }
           });
@@ -835,38 +926,41 @@ export function registerWorktreeHandlers(
               encoding: 'utf-8'
             }).trim();
 
-            // Get base branch
+            // Get base branch - the current branch in the main project (where changes will be merged)
             let baseBranch = 'main';
             try {
-              baseBranch = execSync('git rev-parse --abbrev-ref origin/HEAD 2>/dev/null || echo main', {
+              baseBranch = execSync('git rev-parse --abbrev-ref HEAD', {
                 cwd: project.path,
                 encoding: 'utf-8'
-              }).trim().replace('origin/', '');
+              }).trim();
             } catch {
               baseBranch = 'main';
             }
 
-            // Get commit count
+            // Get commit count (cross-platform - no shell syntax)
             let commitCount = 0;
             try {
-              const countOutput = execSync(`git rev-list --count ${baseBranch}..HEAD 2>/dev/null || echo 0`, {
+              const countOutput = execSync(`git rev-list --count ${baseBranch}..HEAD`, {
                 cwd: entryPath,
-                encoding: 'utf-8'
+                encoding: 'utf-8',
+                stdio: ['pipe', 'pipe', 'pipe']
               }).trim();
               commitCount = parseInt(countOutput, 10) || 0;
             } catch {
               commitCount = 0;
             }
 
-            // Get diff stats
+            // Get diff stats (cross-platform - no shell syntax)
             let filesChanged = 0;
             let additions = 0;
             let deletions = 0;
+            let diffStat = '';
 
             try {
-              const diffStat = execSync(`git diff --shortstat ${baseBranch}...HEAD 2>/dev/null || echo ""`, {
+              diffStat = execSync(`git diff --shortstat ${baseBranch}...HEAD`, {
                 cwd: entryPath,
-                encoding: 'utf-8'
+                encoding: 'utf-8',
+                stdio: ['pipe', 'pipe', 'pipe']
               }).trim();
 
               const filesMatch = diffStat.match(/(\d+) files? changed/);

auto-claude-ui/src/main/ipc-handlers/terminal-handlers.ts

@@ -7,6 +7,8 @@ import { getUsageMonitor } from '../claude-profile/usage-monitor';
 import { TerminalManager } from '../terminal-manager';
 import { projectStore } from '../project-store';
 import { terminalNameGenerator } from '../terminal-name-generator';
+import { debugLog, debugError } from '../../shared/utils/debug-logger';
+import { escapeShellArg, escapeShellArgWindows } from '../../shared/utils/shell-escape';
 
 
 /**
@@ -162,14 +164,108 @@ export function registerTerminalHandlers(
   ipcMain.handle(
     IPC_CHANNELS.CLAUDE_PROFILE_SET_ACTIVE,
     async (_, profileId: string): Promise<IPCResult> => {
+      debugLog('[terminal-handlers:CLAUDE_PROFILE_SET_ACTIVE] ========== PROFILE SWITCH START ==========');
+      debugLog('[terminal-handlers:CLAUDE_PROFILE_SET_ACTIVE] Requested profile ID:', profileId);
+
       try {
         const profileManager = getClaudeProfileManager();
+        const previousProfile = profileManager.getActiveProfile();
+        const previousProfileId = previousProfile.id;
+        const newProfile = profileManager.getProfile(profileId);
+
+        debugLog('[terminal-handlers:CLAUDE_PROFILE_SET_ACTIVE] Previous profile:', {
+          id: previousProfile.id,
+          name: previousProfile.name,
+          hasOAuthToken: !!previousProfile.oauthToken,
+          isDefault: previousProfile.isDefault
+        });
+
+        debugLog('[terminal-handlers:CLAUDE_PROFILE_SET_ACTIVE] New profile:', newProfile ? {
+          id: newProfile.id,
+          name: newProfile.name,
+          hasOAuthToken: !!newProfile.oauthToken,
+          isDefault: newProfile.isDefault
+        } : 'NOT FOUND');
+
         const success = profileManager.setActiveProfile(profileId);
+        debugLog('[terminal-handlers:CLAUDE_PROFILE_SET_ACTIVE] setActiveProfile result:', success);
+
         if (!success) {
+          debugError('[terminal-handlers:CLAUDE_PROFILE_SET_ACTIVE] Profile not found, aborting');
           return { success: false, error: 'Profile not found' };
         }
+
+        // If the profile actually changed, restart Claude in active terminals
+        // This ensures existing Claude sessions use the new profile's OAuth token
+        const profileChanged = previousProfileId !== profileId;
+        debugLog('[terminal-handlers:CLAUDE_PROFILE_SET_ACTIVE] Profile changed:', profileChanged, {
+          previousProfileId,
+          newProfileId: profileId
+        });
+
+        if (profileChanged) {
+          const activeTerminalIds = terminalManager.getActiveTerminalIds();
+          debugLog('[terminal-handlers:CLAUDE_PROFILE_SET_ACTIVE] Active terminal IDs:', activeTerminalIds);
+
+          const switchPromises: Promise<void>[] = [];
+          const terminalsInClaudeMode: string[] = [];
+          const terminalsNotInClaudeMode: string[] = [];
+
+          for (const terminalId of activeTerminalIds) {
+            const isClaudeMode = terminalManager.isClaudeMode(terminalId);
+            debugLog('[terminal-handlers:CLAUDE_PROFILE_SET_ACTIVE] Terminal check:', {
+              terminalId,
+              isClaudeMode
+            });
+
+            if (isClaudeMode) {
+              terminalsInClaudeMode.push(terminalId);
+              debugLog('[terminal-handlers:CLAUDE_PROFILE_SET_ACTIVE] Queuing terminal for profile switch:', terminalId);
+              switchPromises.push(
+                terminalManager.switchClaudeProfile(terminalId, profileId)
+                  .then(() => {
+                    debugLog('[terminal-handlers:CLAUDE_PROFILE_SET_ACTIVE] Terminal profile switch SUCCESS:', terminalId);
+                  })
+                  .catch((err) => {
+                    debugError('[terminal-handlers:CLAUDE_PROFILE_SET_ACTIVE] Terminal profile switch FAILED:', terminalId, err);
+                    throw err; // Re-throw so Promise.allSettled correctly reports rejections
+                  })
+              );
+            } else {
+              terminalsNotInClaudeMode.push(terminalId);
+            }
+          }
+
+          debugLog('[terminal-handlers:CLAUDE_PROFILE_SET_ACTIVE] Terminal summary:', {
+            total: activeTerminalIds.length,
+            inClaudeMode: terminalsInClaudeMode.length,
+            notInClaudeMode: terminalsNotInClaudeMode.length,
+            terminalsToSwitch: terminalsInClaudeMode,
+            terminalsSkipped: terminalsNotInClaudeMode
+          });
+
+          // Wait for all switches to complete (but don't fail the main operation if some fail)
+          if (switchPromises.length > 0) {
+            debugLog('[terminal-handlers:CLAUDE_PROFILE_SET_ACTIVE] Waiting for', switchPromises.length, 'terminal switches...');
+            const results = await Promise.allSettled(switchPromises);
+            const fulfilled = results.filter(r => r.status === 'fulfilled').length;
+            const rejected = results.filter(r => r.status === 'rejected').length;
+            debugLog('[terminal-handlers:CLAUDE_PROFILE_SET_ACTIVE] Switch results:', {
+              total: results.length,
+              fulfilled,
+              rejected
+            });
+          } else {
+            debugLog('[terminal-handlers:CLAUDE_PROFILE_SET_ACTIVE] No terminals in Claude mode to switch');
+          }
+        } else {
+          debugLog('[terminal-handlers:CLAUDE_PROFILE_SET_ACTIVE] Same profile selected, no terminal switches needed');
+        }
+
+        debugLog('[terminal-handlers:CLAUDE_PROFILE_SET_ACTIVE] ========== PROFILE SWITCH COMPLETE ==========');
         return { success: true };
       } catch (error) {
+        debugError('[terminal-handlers:CLAUDE_PROFILE_SET_ACTIVE] EXCEPTION:', error);
         return {
           success: false,
           error: error instanceof Error ? error.message : 'Failed to set active Claude profile'
@@ -208,7 +304,7 @@ export function registerTerminalHandlers(
           const { mkdirSync, existsSync } = await import('fs');
           if (!existsSync(profile.configDir)) {
             mkdirSync(profile.configDir, { recursive: true });
-            console.warn('[IPC] Created config directory:', profile.configDir);
+            debugLog('[IPC] Created config directory:', profile.configDir);
           }
         }
 
@@ -217,7 +313,7 @@ export function registerTerminalHandlers(
         const terminalId = `claude-login-${profileId}-${Date.now()}`;
         const homeDir = process.env.HOME || process.env.USERPROFILE || '/tmp';
 
-        console.warn('[IPC] Initializing Claude profile:', {
+        debugLog('[IPC] Initializing Claude profile:', {
           profileId,
           profileName: profile.name,
           configDir: profile.configDir,
@@ -231,16 +327,25 @@ export function registerTerminalHandlers(
         await new Promise(resolve => setTimeout(resolve, 500));
 
         // Build the login command with the profile's config dir
-        // Use export to ensure the variable persists, then run setup-token
+        // Use platform-specific syntax and escaping for environment variables
         let loginCommand: string;
         if (!profile.isDefault && profile.configDir) {
-          // Use export and run in subshell to ensure CLAUDE_CONFIG_DIR is properly set
-          loginCommand = `export CLAUDE_CONFIG_DIR="${profile.configDir}" && echo "Config dir: $CLAUDE_CONFIG_DIR" && claude setup-token`;
+          if (process.platform === 'win32') {
+            // SECURITY: Use Windows-specific escaping for cmd.exe
+            const escapedConfigDir = escapeShellArgWindows(profile.configDir);
+            // Windows cmd.exe syntax: set "VAR=value" with %VAR% for expansion
+            loginCommand = `set "CLAUDE_CONFIG_DIR=${escapedConfigDir}" && echo Config dir: %CLAUDE_CONFIG_DIR% && claude setup-token`;
+          } else {
+            // SECURITY: Use POSIX escaping for bash/zsh
+            const escapedConfigDir = escapeShellArg(profile.configDir);
+            // Unix/Mac bash/zsh syntax: export VAR=value with $VAR for expansion
+            loginCommand = `export CLAUDE_CONFIG_DIR=${escapedConfigDir} && echo "Config dir: $CLAUDE_CONFIG_DIR" && claude setup-token`;
+          }
         } else {
           loginCommand = 'claude setup-token';
         }
 
-        console.warn('[IPC] Sending login command to terminal:', loginCommand);
+        debugLog('[IPC] Sending login command to terminal:', loginCommand);
 
         // Write the login command to the terminal
         terminalManager.write(terminalId, `${loginCommand}\r`);
@@ -263,7 +368,7 @@ export function registerTerminalHandlers(
           }
         };
       } catch (error) {
-        console.error('[IPC] Failed to initialize Claude profile:', error);
+        debugError('[IPC] Failed to initialize Claude profile:', error);
         return {
           success: false,
           error: error instanceof Error ? error.message : 'Failed to initialize Claude profile'
@@ -284,7 +389,7 @@ export function registerTerminalHandlers(
         }
         return { success: true };
       } catch (error) {
-        console.error('[IPC] Failed to set OAuth token:', error);
+        debugError('[IPC] Failed to set OAuth token:', error);
         return {
           success: false,
           error: error instanceof Error ? error.message : 'Failed to set OAuth token'
@@ -569,5 +674,5 @@ export function initializeUsageMonitorForwarding(mainWindow: BrowserWindow): voi
     mainWindow.webContents.send(IPC_CHANNELS.PROACTIVE_SWAP_NOTIFICATION, notification);
   });
 
-  console.warn('[terminal-handlers] Usage monitor event forwarding initialized');
+  debugLog('[terminal-handlers] Usage monitor event forwarding initialized');
 }

auto-claude-ui/src/main/memory-service.ts

@@ -0,0 +1,631 @@
+/**
+ * Memory Service
+ *
+ * Queries the LadybugDB graph database for memories stored by Graphiti.
+ * Uses Python subprocess to communicate with the embedded database.
+ *
+ * LadybugDB stores data in Kuzu format at ~/.auto-claude/memories/<database>/
+ */
+
+import { spawn } from 'child_process';
+import * as path from 'path';
+import * as os from 'os';
+import * as fs from 'fs';
+import { app } from 'electron';
+import { findPythonCommand, parsePythonCommand } from './python-detector';
+import type { MemoryEpisode } from '../shared/types';
+
+interface MemoryServiceConfig {
+  dbPath: string;
+  database: string;
+}
+
+// Embedder configuration for semantic search
+export interface EmbedderConfig {
+  provider: 'openai' | 'google' | 'ollama' | 'voyage' | 'azure_openai';
+  // OpenAI
+  openaiApiKey?: string;
+  openaiEmbeddingModel?: string;
+  // Google AI
+  googleApiKey?: string;
+  googleEmbeddingModel?: string;
+  // Ollama
+  ollamaBaseUrl?: string;
+  ollamaEmbeddingModel?: string;
+  ollamaEmbeddingDim?: number;
+  // Voyage AI
+  voyageApiKey?: string;
+  voyageEmbeddingModel?: string;
+  // Azure OpenAI
+  azureOpenaiApiKey?: string;
+  azureOpenaiBaseUrl?: string;
+  azureOpenaiEmbeddingDeployment?: string;
+}
+
+interface SemanticSearchResult extends MemoryQueryResult {
+  search_type: 'semantic' | 'keyword';
+  embedder?: string;
+}
+
+interface QueryResult {
+  success: boolean;
+  data?: unknown;
+  error?: string;
+}
+
+interface MemoryQueryResult {
+  memories: Array<{
+    id: string;
+    name: string;
+    type: string;
+    timestamp: string;
+    content: string;
+    description?: string;
+    group_id?: string;
+    session_number?: number;
+    score?: number;
+  }>;
+  count: number;
+  query?: string;
+}
+
+interface StatusResult {
+  available: boolean;
+  ladybugInstalled: boolean;
+  databasePath: string;
+  database: string;
+  databaseExists: boolean;
+  connected?: boolean;
+  databases?: string[];
+  error?: string | null;
+}
+
+/**
+ * Get the default database path
+ */
+export function getDefaultDbPath(): string {
+  return path.join(os.homedir(), '.auto-claude', 'memories');
+}
+
+/**
+ * Get the path to the query_memory.py script
+ */
+function getQueryScriptPath(): string | null {
+  // Look for the script in auto-claude directory (sibling to auto-claude-ui)
+  const possiblePaths = [
+    // Dev mode: from dist/main -> ../../auto-claude
+    path.resolve(__dirname, '..', '..', '..', 'auto-claude', 'query_memory.py'),
+    // Packaged app: from app.getAppPath() (handles asar and resources correctly)
+    path.resolve(app.getAppPath(), '..', 'auto-claude', 'query_memory.py'),
+    // Alternative: from app root
+    path.resolve(process.cwd(), 'auto-claude', 'query_memory.py'),
+    // If running from repo root
+    path.resolve(process.cwd(), '..', 'auto-claude', 'query_memory.py'),
+  ];
+
+  for (const p of possiblePaths) {
+    if (fs.existsSync(p)) {
+      return p;
+    }
+  }
+  return null;
+}
+
+/**
+ * Execute a Python memory query command
+ */
+async function executeQuery(
+  command: string,
+  args: string[],
+  timeout: number = 10000
+): Promise<QueryResult> {
+  const pythonCmd = findPythonCommand();
+  if (!pythonCmd) {
+    return { success: false, error: 'Python not found' };
+  }
+
+  const scriptPath = getQueryScriptPath();
+  if (!scriptPath) {
+    return { success: false, error: 'query_memory.py script not found' };
+  }
+
+  const [pythonExe, baseArgs] = parsePythonCommand(pythonCmd);
+
+  return new Promise((resolve) => {
+    const fullArgs = [...baseArgs, scriptPath, command, ...args];
+    const proc = spawn(pythonExe, fullArgs, {
+      stdio: ['ignore', 'pipe', 'pipe'],
+      timeout,
+    });
+
+    let stdout = '';
+    let stderr = '';
+
+    proc.stdout.on('data', (data) => {
+      stdout += data.toString();
+    });
+
+    proc.stderr.on('data', (data) => {
+      stderr += data.toString();
+    });
+
+    proc.on('close', (code) => {
+      if (code === 0 && stdout) {
+        try {
+          const result = JSON.parse(stdout);
+          resolve(result);
+        } catch {
+          resolve({ success: false, error: `Invalid JSON response: ${stdout}` });
+        }
+      } else {
+        resolve({
+          success: false,
+          error: stderr || `Process exited with code ${code}`,
+        });
+      }
+    });
+
+    proc.on('error', (err) => {
+      resolve({ success: false, error: err.message });
+    });
+
+    // Handle timeout
+    setTimeout(() => {
+      proc.kill();
+      resolve({ success: false, error: 'Query timed out' });
+    }, timeout);
+  });
+}
+
+/**
+ * Execute semantic search with embedder configuration passed via environment
+ */
+async function executeSemanticQuery(
+  args: string[],
+  embedderConfig: EmbedderConfig,
+  timeout: number = 30000 // Longer timeout for embedding operations
+): Promise<QueryResult> {
+  const pythonCmd = findPythonCommand();
+  if (!pythonCmd) {
+    return { success: false, error: 'Python not found' };
+  }
+
+  const scriptPath = getQueryScriptPath();
+  if (!scriptPath) {
+    return { success: false, error: 'query_memory.py script not found' };
+  }
+
+  const [pythonExe, baseArgs] = parsePythonCommand(pythonCmd);
+
+  // Build environment with embedder configuration
+  const env: Record<string, string | undefined> = { ...process.env };
+
+  // Set the embedder provider
+  env.GRAPHITI_EMBEDDER_PROVIDER = embedderConfig.provider;
+
+  // Provider-specific configuration
+  switch (embedderConfig.provider) {
+    case 'openai':
+      if (embedderConfig.openaiApiKey) {
+        env.OPENAI_API_KEY = embedderConfig.openaiApiKey;
+      }
+      if (embedderConfig.openaiEmbeddingModel) {
+        env.OPENAI_EMBEDDING_MODEL = embedderConfig.openaiEmbeddingModel;
+      }
+      break;
+
+    case 'google':
+      if (embedderConfig.googleApiKey) {
+        env.GOOGLE_API_KEY = embedderConfig.googleApiKey;
+      }
+      if (embedderConfig.googleEmbeddingModel) {
+        env.GOOGLE_EMBEDDING_MODEL = embedderConfig.googleEmbeddingModel;
+      }
+      break;
+
+    case 'ollama':
+      if (embedderConfig.ollamaBaseUrl) {
+        env.OLLAMA_BASE_URL = embedderConfig.ollamaBaseUrl;
+      }
+      if (embedderConfig.ollamaEmbeddingModel) {
+        env.OLLAMA_EMBEDDING_MODEL = embedderConfig.ollamaEmbeddingModel;
+      }
+      if (embedderConfig.ollamaEmbeddingDim) {
+        env.OLLAMA_EMBEDDING_DIM = String(embedderConfig.ollamaEmbeddingDim);
+      }
+      break;
+
+    case 'voyage':
+      if (embedderConfig.voyageApiKey) {
+        env.VOYAGE_API_KEY = embedderConfig.voyageApiKey;
+      }
+      if (embedderConfig.voyageEmbeddingModel) {
+        env.VOYAGE_EMBEDDING_MODEL = embedderConfig.voyageEmbeddingModel;
+      }
+      break;
+
+    case 'azure_openai':
+      if (embedderConfig.azureOpenaiApiKey) {
+        env.AZURE_OPENAI_API_KEY = embedderConfig.azureOpenaiApiKey;
+      }
+      if (embedderConfig.azureOpenaiBaseUrl) {
+        env.AZURE_OPENAI_BASE_URL = embedderConfig.azureOpenaiBaseUrl;
+      }
+      if (embedderConfig.azureOpenaiEmbeddingDeployment) {
+        env.AZURE_OPENAI_EMBEDDING_DEPLOYMENT = embedderConfig.azureOpenaiEmbeddingDeployment;
+      }
+      break;
+  }
+
+  return new Promise((resolve) => {
+    const fullArgs = [...baseArgs, scriptPath, 'semantic-search', ...args];
+    const proc = spawn(pythonExe, fullArgs, {
+      stdio: ['ignore', 'pipe', 'pipe'],
+      env,
+      timeout,
+    });
+
+    let stdout = '';
+    let stderr = '';
+
+    proc.stdout.on('data', (data) => {
+      stdout += data.toString();
+    });
+
+    proc.stderr.on('data', (data) => {
+      stderr += data.toString();
+    });
+
+    proc.on('close', (code) => {
+      if (code === 0 && stdout) {
+        try {
+          const result = JSON.parse(stdout);
+          resolve(result);
+        } catch {
+          resolve({ success: false, error: `Invalid JSON response: ${stdout}` });
+        }
+      } else {
+        resolve({
+          success: false,
+          error: stderr || `Process exited with code ${code}`,
+        });
+      }
+    });
+
+    proc.on('error', (err) => {
+      resolve({ success: false, error: err.message });
+    });
+
+    setTimeout(() => {
+      proc.kill();
+      resolve({ success: false, error: 'Semantic search timed out' });
+    }, timeout);
+  });
+}
+
+/**
+ * Memory Service for querying graph memories from LadybugDB
+ */
+export class MemoryService {
+  private config: MemoryServiceConfig;
+
+  constructor(config: MemoryServiceConfig) {
+    this.config = config;
+  }
+
+  /**
+   * Get the full path to the database
+   */
+  private getDbFullPath(): string {
+    return path.join(this.config.dbPath, this.config.database);
+  }
+
+  /**
+   * Check if the database exists
+   */
+  databaseExists(): boolean {
+    const dbPath = this.getDbFullPath();
+    return fs.existsSync(dbPath);
+  }
+
+  /**
+   * List all available databases
+   */
+  listDatabases(): string[] {
+    try {
+      const basePath = this.config.dbPath;
+      if (!fs.existsSync(basePath)) {
+        return [];
+      }
+
+      return fs.readdirSync(basePath).filter((name) => {
+        if (name.startsWith('.')) return false;
+        return true; // Include both files and directories
+      });
+    } catch (error) {
+      console.error('Failed to list databases:', error);
+      return [];
+    }
+  }
+
+  /**
+   * Query episodic memories from the database
+   */
+  async getEpisodicMemories(limit: number = 20): Promise<MemoryEpisode[]> {
+    const result = await executeQuery('get-memories', [
+      this.config.dbPath,
+      this.config.database,
+      '--limit',
+      String(limit),
+    ]);
+
+    if (!result.success || !result.data) {
+      console.error('Failed to get memories:', result.error);
+      return [];
+    }
+
+    const data = result.data as MemoryQueryResult;
+    return data.memories.map((m) => ({
+      id: m.id,
+      type: this.mapMemoryType(m.type),
+      timestamp: m.timestamp,
+      content: m.content,
+      session_number: m.session_number,
+    }));
+  }
+
+  /**
+   * Query entity memories (patterns, gotchas, etc.) from the database
+   */
+  async getEntityMemories(limit: number = 20): Promise<MemoryEpisode[]> {
+    const result = await executeQuery('get-entities', [
+      this.config.dbPath,
+      this.config.database,
+      '--limit',
+      String(limit),
+    ]);
+
+    if (!result.success || !result.data) {
+      console.error('Failed to get entities:', result.error);
+      return [];
+    }
+
+    const data = result.data as { entities: MemoryQueryResult['memories']; count: number };
+    return data.entities.map((e) => ({
+      id: e.id,
+      type: this.mapMemoryType(e.type),
+      timestamp: e.timestamp,
+      content: e.content,
+    }));
+  }
+
+  /**
+   * Get all memories from the database
+   */
+  async getAllMemories(limit: number = 20): Promise<MemoryEpisode[]> {
+    const [episodic, entities] = await Promise.all([
+      this.getEpisodicMemories(limit),
+      this.getEntityMemories(limit),
+    ]);
+
+    const memories = [...episodic, ...entities];
+
+    // Sort by timestamp descending
+    memories.sort((a, b) => new Date(b.timestamp).getTime() - new Date(a.timestamp).getTime());
+
+    return memories.slice(0, limit);
+  }
+
+  /**
+   * Search memories in the database (keyword search)
+   */
+  async searchMemories(searchQuery: string, limit: number = 20): Promise<MemoryEpisode[]> {
+    const result = await executeQuery('search', [
+      this.config.dbPath,
+      this.config.database,
+      searchQuery,
+      '--limit',
+      String(limit),
+    ]);
+
+    if (!result.success || !result.data) {
+      console.error('Failed to search memories:', result.error);
+      return [];
+    }
+
+    const data = result.data as MemoryQueryResult;
+    return data.memories.map((m) => ({
+      id: m.id,
+      type: this.mapMemoryType(m.type),
+      timestamp: m.timestamp,
+      content: m.content,
+      session_number: m.session_number,
+      score: m.score,
+    }));
+  }
+
+  /**
+   * Semantic search using embeddings
+   *
+   * Uses the configured embedder to create vector embeddings and perform
+   * similarity search. Falls back to keyword search if embedder fails.
+   *
+   * @param searchQuery The search query
+   * @param embedderConfig Configuration for the embedding provider
+   * @param limit Maximum number of results
+   * @returns Memories with relevance scores
+   */
+  async searchMemoriesSemantic(
+    searchQuery: string,
+    embedderConfig: EmbedderConfig,
+    limit: number = 20
+  ): Promise<{ memories: MemoryEpisode[]; searchType: 'semantic' | 'keyword' }> {
+    const result = await executeSemanticQuery(
+      [this.config.dbPath, this.config.database, searchQuery, '--limit', String(limit)],
+      embedderConfig
+    );
+
+    if (!result.success || !result.data) {
+      console.error('Semantic search failed, falling back to keyword:', result.error);
+      // Fall back to keyword search
+      const memories = await this.searchMemories(searchQuery, limit);
+      return { memories, searchType: 'keyword' };
+    }
+
+    const data = result.data as SemanticSearchResult;
+    const memories = data.memories.map((m) => ({
+      id: m.id,
+      type: this.mapMemoryType(m.type),
+      timestamp: m.timestamp,
+      content: m.content,
+      session_number: m.session_number,
+      score: m.score,
+    }));
+
+    return {
+      memories,
+      searchType: data.search_type || 'semantic',
+    };
+  }
+
+  /**
+   * Test connection to the database
+   */
+  async testConnection(): Promise<{ success: boolean; message: string }> {
+    const result = await executeQuery('get-status', [this.config.dbPath, this.config.database]);
+
+    if (!result.success) {
+      return {
+        success: false,
+        message: result.error || 'Failed to check database status',
+      };
+    }
+
+    const data = result.data as StatusResult;
+
+    if (!data.available) {
+      return {
+        success: false,
+        message: 'LadybugDB (real_ladybug) not installed. Requires Python 3.12+',
+      };
+    }
+
+    if (!data.databaseExists) {
+      return {
+        success: false,
+        message: `Database not found at ${data.databasePath}/${data.database}`,
+      };
+    }
+
+    if (!data.connected) {
+      return {
+        success: false,
+        message: data.error || 'Failed to connect to database',
+      };
+    }
+
+    const dbCount = data.databases?.length || 0;
+    return {
+      success: true,
+      message: `Connected to LadybugDB with ${dbCount} databases`,
+    };
+  }
+
+  /**
+   * Close the database connection (no-op for subprocess model)
+   */
+  async close(): Promise<void> {
+    // No persistent connection to close with subprocess model
+  }
+
+  /**
+   * Map string type to MemoryEpisode type
+   */
+  private mapMemoryType(type: string): MemoryEpisode['type'] {
+    switch (type) {
+      case 'session_insight':
+        return 'session_insight';
+      case 'pattern':
+        return 'pattern';
+      case 'gotcha':
+        return 'gotcha';
+      case 'codebase_discovery':
+        return 'codebase_discovery';
+      case 'task_outcome':
+        return 'task_outcome';
+      default:
+        return 'session_insight';
+    }
+  }
+}
+
+// Singleton instance for reuse
+let serviceInstance: MemoryService | null = null;
+
+/**
+ * Get or create a Memory service instance
+ */
+export function getMemoryService(config: MemoryServiceConfig): MemoryService {
+  if (
+    !serviceInstance ||
+    serviceInstance['config'].dbPath !== config.dbPath ||
+    serviceInstance['config'].database !== config.database
+  ) {
+    serviceInstance = new MemoryService(config);
+  }
+  return serviceInstance;
+}
+
+/**
+ * Close the singleton service instance
+ */
+export async function closeMemoryService(): Promise<void> {
+  if (serviceInstance) {
+    await serviceInstance.close();
+    serviceInstance = null;
+  }
+}
+
+/**
+ * Check if Python with LadybugDB is available
+ */
+export function isKuzuAvailable(): boolean {
+  // Check if Python is available
+  const pythonCmd = findPythonCommand();
+  if (!pythonCmd) {
+    return false;
+  }
+
+  // Check if query script exists
+  const scriptPath = getQueryScriptPath();
+  return scriptPath !== null;
+}
+
+/**
+ * Get memory service status
+ */
+export interface MemoryServiceStatus {
+  kuzuInstalled: boolean;
+  databasePath: string;
+  databaseExists: boolean;
+  databases: string[];
+}
+
+export function getMemoryServiceStatus(dbPath?: string): MemoryServiceStatus {
+  const basePath = dbPath || getDefaultDbPath();
+
+  const databases = fs.existsSync(basePath)
+    ? fs.readdirSync(basePath).filter((name) => !name.startsWith('.'))
+    : [];
+
+  // Check if Python and script are available
+  const pythonAvailable = findPythonCommand() !== null;
+  const scriptAvailable = getQueryScriptPath() !== null;
+
+  return {
+    kuzuInstalled: pythonAvailable && scriptAvailable,
+    databasePath: basePath,
+    databaseExists: databases.length > 0,
+    databases,
+  };
+}

auto-claude-ui/src/main/notification-service.ts

@@ -66,7 +66,7 @@ class NotificationService {
   private sendNotification(type: NotificationType, options: NotificationOptions): void {
     // Get notification settings
     const settings = this.getNotificationSettings(options.projectId);
-    
+
     // Check if this notification type is enabled
     if (!this.isNotificationEnabled(type, settings)) {
       return;
@@ -162,4 +162,3 @@ class NotificationService {
 
 // Export singleton instance
 export const notificationService = new NotificationService();
-

auto-claude-ui/src/main/project-initializer.ts

@@ -3,9 +3,9 @@ import path from 'path';
 import { execSync } from 'child_process';
 
 /**
- * Debug logging - only logs when AUTO_CLAUDE_DEBUG env var is set
+ * Debug logging - only logs when DEBUG=true or in development mode
  */
-const DEBUG = process.env.AUTO_CLAUDE_DEBUG === 'true' || process.env.AUTO_CLAUDE_DEBUG === '1';
+const DEBUG = process.env.DEBUG === 'true' || process.env.NODE_ENV === 'development';
 
 function debug(message: string, data?: Record<string, unknown>): void {
   if (DEBUG) {

auto-claude-ui/src/main/project-store.ts

@@ -6,9 +6,16 @@ import type { Project, ProjectSettings, Task, TaskStatus, TaskMetadata, Implemen
 import { DEFAULT_PROJECT_SETTINGS, AUTO_BUILD_PATHS, getSpecsDir } from '../shared/constants';
 import { getAutoBuildPath, isInitialized } from './project-initializer';
 
+interface TabState {
+  openProjectIds: string[];
+  activeProjectId: string | null;
+  tabOrder: string[];
+}
+
 interface StoreData {
   projects: Project[];
   settings: Record<string, unknown>;
+  tabState?: TabState;
 }
 
 /**
@@ -68,6 +75,14 @@ export class ProjectStore {
     // Check if project already exists
     const existing = this.data.projects.find((p) => p.path === projectPath);
     if (existing) {
+      // Validate that .auto-claude folder still exists for existing project
+      // If manually deleted, reset autoBuildPath so UI prompts for reinitialization
+      if (existing.autoBuildPath && !isInitialized(existing.path)) {
+        console.warn(`[ProjectStore] .auto-claude folder was deleted for project "${existing.name}" - resetting autoBuildPath`);
+        existing.autoBuildPath = '';
+        existing.updatedAt = new Date();
+        this.save();
+      }
       return existing;
     }
 
@@ -126,6 +141,34 @@ export class ProjectStore {
     return this.data.projects;
   }
 
+  /**
+   * Get tab state
+   */
+  getTabState(): TabState {
+    return this.data.tabState || {
+      openProjectIds: [],
+      activeProjectId: null,
+      tabOrder: []
+    };
+  }
+
+  /**
+   * Save tab state
+   */
+  saveTabState(tabState: TabState): void {
+    // Filter out any project IDs that no longer exist
+    const validProjectIds = this.data.projects.map(p => p.id);
+    this.data.tabState = {
+      openProjectIds: tabState.openProjectIds.filter(id => validProjectIds.includes(id)),
+      activeProjectId: tabState.activeProjectId && validProjectIds.includes(tabState.activeProjectId)
+        ? tabState.activeProjectId
+        : null,
+      tabOrder: tabState.tabOrder.filter(id => validProjectIds.includes(id))
+    };
+    console.log('[ProjectStore] Saving tab state:', this.data.tabState);
+    this.save();
+  }
+
   /**
    * Validate all projects to ensure their .auto-claude folders still exist.
    * If a project has autoBuildPath set but the folder was deleted,
@@ -243,8 +286,8 @@ export class ProjectStore {
         if (existsSync(specFilePath)) {
           try {
             const content = readFileSync(specFilePath, 'utf-8');
-            // Extract first paragraph after "## Overview"
-            const overviewMatch = content.match(/## Overview\s*\n\n([^\n#]+)/);
+            // Extract first paragraph after "## Overview" - handle both with and without blank line
+            const overviewMatch = content.match(/## Overview\s*\n+([^\n#]+)/);
             if (overviewMatch) {
               description = overviewMatch[1].trim();
             }
@@ -258,6 +301,42 @@ export class ProjectStore {
           description = plan.description;
         }
 
+        // Fallback: read description from requirements.json if still not found
+        if (!description) {
+          const requirementsPath = path.join(specPath, AUTO_BUILD_PATHS.REQUIREMENTS);
+          if (existsSync(requirementsPath)) {
+            try {
+              const reqContent = readFileSync(requirementsPath, 'utf-8');
+              const requirements = JSON.parse(reqContent);
+              if (requirements.task_description) {
+                // Extract a clean summary from task_description (first line or first ~200 chars)
+                const taskDesc = requirements.task_description;
+                const firstLine = taskDesc.split('\n')[0].trim();
+                // If the first line is a title like "Investigate GitHub Issue #36", use the next meaningful line
+                if (firstLine.toLowerCase().startsWith('investigate') && taskDesc.includes('\n\n')) {
+                  const sections = taskDesc.split('\n\n');
+                  // Find the first paragraph that's not a title
+                  for (const section of sections) {
+                    const trimmed = section.trim();
+                    // Skip headers and short lines
+                    if (trimmed.startsWith('#') || trimmed.length < 20) continue;
+                    // Skip the "Please analyze" instruction at the end
+                    if (trimmed.startsWith('Please analyze')) continue;
+                    description = trimmed.substring(0, 200).split('\n')[0];
+                    break;
+                  }
+                }
+                // If still no description, use a shortened version of task_description
+                if (!description) {
+                  description = firstLine.substring(0, 150);
+                }
+              }
+            } catch {
+              // Ignore parse errors
+            }
+          }
+        }
+
         // Try to read task metadata
         const metadataPath = path.join(specPath, 'task_metadata.json');
         let metadata: TaskMetadata | undefined;
@@ -290,11 +369,30 @@ export class ProjectStore {
         const stagedInMainProject = planWithStaged?.stagedInMainProject;
         const stagedAt = planWithStaged?.stagedAt;
 
+        // Determine title - check if feature looks like a spec ID (e.g., "054-something-something")
+        let title = plan?.feature || plan?.title || dir.name;
+        const looksLikeSpecId = /^\d{3}-/.test(title);
+        if (looksLikeSpecId && existsSync(specFilePath)) {
+          try {
+            const specContent = readFileSync(specFilePath, 'utf-8');
+            // Extract title from first # line, handling patterns like:
+            // "# Quick Spec: Title" -> "Title"
+            // "# Specification: Title" -> "Title"
+            // "# Title" -> "Title"
+            const titleMatch = specContent.match(/^#\s+(?:Quick Spec:|Specification:)?\s*(.+)$/m);
+            if (titleMatch && titleMatch[1]) {
+              title = titleMatch[1].trim();
+            }
+          } catch {
+            // Keep the original title on error
+          }
+        }
+
         tasks.push({
           id: dir.name, // Use spec directory name as ID
           specId: dir.name,
           projectId,
-          title: plan?.feature || plan?.title || dir.name,
+          title,
           description,
           status,
           reviewReason,

auto-claude-ui/src/main/python-detector.ts

@@ -0,0 +1,61 @@
+import { execSync } from 'child_process';
+
+/**
+ * Detect and return the best available Python command.
+ * Tries multiple candidates and returns the first one that works with Python 3.
+ *
+ * @returns The Python command to use, or null if none found
+ */
+export function findPythonCommand(): string | null {
+  const isWindows = process.platform === 'win32';
+
+  // On Windows, try py launcher first (most reliable), then python, then python3
+  // On Unix, try python3 first, then python
+  const candidates = isWindows
+    ? ['py -3', 'python', 'python3', 'py']
+    : ['python3', 'python'];
+
+  for (const cmd of candidates) {
+    try {
+      const version = execSync(`${cmd} --version`, {
+        stdio: 'pipe',
+        timeout: 5000,
+        windowsHide: true
+      }).toString();
+
+      if (version.includes('Python 3')) {
+        return cmd;
+      }
+    } catch {
+      // Command not found or errored, try next
+      continue;
+    }
+  }
+
+  // Fallback to platform-specific default
+  return isWindows ? 'python' : 'python3';
+}
+
+/**
+ * Get the default Python command for the current platform.
+ * This is a synchronous fallback that doesn't test if Python actually exists.
+ *
+ * @returns The default Python command for this platform
+ */
+export function getDefaultPythonCommand(): string {
+  return process.platform === 'win32' ? 'python' : 'python3';
+}
+
+/**
+ * Parse a Python command string into command and base arguments.
+ * Handles space-separated commands like "py -3".
+ *
+ * @param pythonPath - The Python command string (e.g., "python3", "py -3")
+ * @returns Tuple of [command, baseArgs] ready for use with spawn()
+ */
+export function parsePythonCommand(pythonPath: string): [string, string[]] {
+  const parts = pythonPath.split(' ');
+  const command = parts[0];
+  const baseArgs = parts.slice(1);
+  return [command, baseArgs];
+}

auto-claude-ui/src/main/python-env-manager.ts

@@ -2,6 +2,7 @@ import { spawn, execSync } from 'child_process';
 import { existsSync } from 'fs';
 import path from 'path';
 import { EventEmitter } from 'events';
+import { app } from 'electron';
 
 export interface PythonEnvStatus {
   ready: boolean;
@@ -14,6 +15,9 @@ export interface PythonEnvStatus {
 /**
  * Manages the Python virtual environment for the auto-claude backend.
  * Automatically creates venv and installs dependencies if needed.
+ *
+ * On packaged apps (especially Linux AppImages), the bundled source is read-only,
+ * so we create the venv in userData instead of inside the source directory.
  */
 export class PythonEnvManager extends EventEmitter {
   private autoBuildSourcePath: string | null = null;
@@ -21,32 +25,46 @@ export class PythonEnvManager extends EventEmitter {
   private isInitializing = false;
   private isReady = false;
 
+  /**
+   * Get the path where the venv should be created.
+   * For packaged apps, this is in userData to avoid read-only filesystem issues.
+   * For development, this is inside the source directory.
+   */
+  private getVenvBasePath(): string | null {
+    if (!this.autoBuildSourcePath) return null;
+
+    // For packaged apps, put venv in userData (writable location)
+    // This fixes Linux AppImage where resources are read-only
+    if (app.isPackaged) {
+      return path.join(app.getPath('userData'), 'python-venv');
+    }
+
+    // Development mode - use source directory
+    return path.join(this.autoBuildSourcePath, '.venv');
+  }
+
   /**
    * Get the path to the venv Python executable
    */
   private getVenvPythonPath(): string | null {
-    if (!this.autoBuildSourcePath) return null;
+    const venvPath = this.getVenvBasePath();
+    if (!venvPath) return null;
 
     const venvPython =
       process.platform === 'win32'
-        ? path.join(this.autoBuildSourcePath, '.venv', 'Scripts', 'python.exe')
-        : path.join(this.autoBuildSourcePath, '.venv', 'bin', 'python');
+        ? path.join(venvPath, 'Scripts', 'python.exe')
+        : path.join(venvPath, 'bin', 'python');
 
     return venvPython;
   }
 
   /**
    * Get the path to pip in the venv
+   * Returns null - we use python -m pip instead for better compatibility
+   * @deprecated Use getVenvPythonPath() with -m pip instead
    */
   private getVenvPipPath(): string | null {
-    if (!this.autoBuildSourcePath) return null;
-
-    const venvPip =
-      process.platform === 'win32'
-        ? path.join(this.autoBuildSourcePath, '.venv', 'Scripts', 'pip.exe')
-        : path.join(this.autoBuildSourcePath, '.venv', 'bin', 'pip');
-
-    return venvPip;
+    return null; // Not used - we use python -m pip
   }
 
   /**
@@ -147,10 +165,10 @@ export class PythonEnvManager extends EventEmitter {
     }
 
     this.emit('status', 'Creating Python virtual environment...');
-    console.warn('[PythonEnvManager] Creating venv with:', systemPython);
+    const venvPath = this.getVenvBasePath()!;
+    console.warn('[PythonEnvManager] Creating venv at:', venvPath, 'with:', systemPython);
 
     return new Promise((resolve) => {
-      const venvPath = path.join(this.autoBuildSourcePath!, '.venv');
       const proc = spawn(systemPython, ['-m', 'venv', venvPath], {
         cwd: this.autoBuildSourcePath!,
         stdio: 'pipe'
@@ -181,16 +199,54 @@ export class PythonEnvManager extends EventEmitter {
   }
 
   /**
-   * Install dependencies from requirements.txt
+   * Bootstrap pip in the venv using ensurepip
+   */
+  private async bootstrapPip(): Promise<boolean> {
+    const venvPython = this.getVenvPythonPath();
+    if (!venvPython || !existsSync(venvPython)) {
+      return false;
+    }
+
+    console.warn('[PythonEnvManager] Bootstrapping pip...');
+    return new Promise((resolve) => {
+      const proc = spawn(venvPython, ['-m', 'ensurepip'], {
+        cwd: this.autoBuildSourcePath!,
+        stdio: 'pipe'
+      });
+
+      let stderr = '';
+      proc.stderr?.on('data', (data) => {
+        stderr += data.toString();
+      });
+
+      proc.on('close', (code) => {
+        if (code === 0) {
+          console.warn('[PythonEnvManager] Pip bootstrapped successfully');
+          resolve(true);
+        } else {
+          console.error('[PythonEnvManager] Failed to bootstrap pip:', stderr);
+          resolve(false);
+        }
+      });
+
+      proc.on('error', (err) => {
+        console.error('[PythonEnvManager] Error bootstrapping pip:', err);
+        resolve(false);
+      });
+    });
+  }
+
+  /**
+   * Install dependencies from requirements.txt using python -m pip
    */
   private async installDeps(): Promise<boolean> {
     if (!this.autoBuildSourcePath) return false;
 
-    const venvPip = this.getVenvPipPath();
+    const venvPython = this.getVenvPythonPath();
     const requirementsPath = path.join(this.autoBuildSourcePath, 'requirements.txt');
 
-    if (!venvPip || !existsSync(venvPip)) {
-      this.emit('error', 'Pip not found in virtual environment');
+    if (!venvPython || !existsSync(venvPython)) {
+      this.emit('error', 'Python not found in virtual environment');
       return false;
     }
 
@@ -199,11 +255,15 @@ export class PythonEnvManager extends EventEmitter {
       return false;
     }
 
+    // Bootstrap pip first if needed
+    await this.bootstrapPip();
+
     this.emit('status', 'Installing Python dependencies (this may take a minute)...');
     console.warn('[PythonEnvManager] Installing dependencies from:', requirementsPath);
 
     return new Promise((resolve) => {
-      const proc = spawn(venvPip, ['install', '-r', requirementsPath], {
+      // Use python -m pip for better compatibility across Python versions
+      const proc = spawn(venvPython, ['-m', 'pip', 'install', '-r', requirementsPath], {
         cwd: this.autoBuildSourcePath!,
         stdio: 'pipe'
       });

auto-claude-ui/src/main/task-log-service.ts

@@ -44,6 +44,7 @@ export class TaskLogService extends EventEmitter {
     try {
       const content = readFileSync(logFile, 'utf-8');
       const logs = JSON.parse(content) as TaskLogs;
+      this.logCache.set(specDir, logs);
       return logs;
     } catch (error) {
       // JSON parse error - file may be mid-write, return cached version if available

auto-claude-ui/src/main/terminal-name-generator.ts

@@ -4,11 +4,13 @@ import { spawn } from 'child_process';
 import { app } from 'electron';
 import { EventEmitter } from 'events';
 import { detectRateLimit, createSDKRateLimitInfo, getProfileEnv } from './rate-limit-detector';
+import { parsePythonCommand } from './python-detector';
+import { pythonEnvManager } from './python-env-manager';
 
 /**
- * Debug logging - only logs when AUTO_CLAUDE_DEBUG env var is set
+ * Debug logging - only logs when DEBUG=true or in development mode
  */
-const DEBUG = process.env.AUTO_CLAUDE_DEBUG === 'true' || process.env.AUTO_CLAUDE_DEBUG === '1';
+const DEBUG = process.env.DEBUG === 'true' || process.env.NODE_ENV === 'development';
 
 function debug(...args: unknown[]): void {
   if (DEBUG) {
@@ -20,7 +22,6 @@ function debug(...args: unknown[]): void {
  * Service for generating terminal names from commands using Claude AI
  */
 export class TerminalNameGenerator extends EventEmitter {
-  private pythonPath: string = 'python3';
   private autoBuildSourcePath: string = '';
 
   constructor() {
@@ -29,12 +30,9 @@ export class TerminalNameGenerator extends EventEmitter {
   }
 
   /**
-   * Configure paths for Python and auto-claude source
+   * Configure the auto-claude source path
    */
-  configure(pythonPath?: string, autoBuildSourcePath?: string): void {
-    if (pythonPath) {
-      this.pythonPath = pythonPath;
-    }
+  configure(autoBuildSourcePath?: string): void {
     if (autoBuildSourcePath) {
       this.autoBuildSourcePath = autoBuildSourcePath;
     }
@@ -116,6 +114,23 @@ export class TerminalNameGenerator extends EventEmitter {
       return null;
     }
 
+    // Check if Python environment is ready (has claude_agent_sdk installed)
+    if (!pythonEnvManager.isEnvReady()) {
+      debug('Python environment not ready, initializing...');
+      const status = await pythonEnvManager.initialize(autoBuildSource);
+      if (!status.ready) {
+        debug('Python environment initialization failed:', status.error);
+        return null;
+      }
+    }
+
+    // Get the venv Python path (where claude_agent_sdk is installed)
+    const venvPythonPath = pythonEnvManager.getPythonPath();
+    if (!venvPythonPath) {
+      debug('Venv Python path not available');
+      return null;
+    }
+
     const prompt = this.createNamePrompt(command, cwd);
     const script = this.createGenerationScript(prompt);
 
@@ -130,7 +145,9 @@ export class TerminalNameGenerator extends EventEmitter {
     const profileEnv = getProfileEnv();
 
     return new Promise((resolve) => {
-      const childProcess = spawn(this.pythonPath, ['-c', script], {
+      // Use the venv Python where claude_agent_sdk is installed
+      const [pythonCommand, pythonBaseArgs] = parsePythonCommand(venvPythonPath);
+      const childProcess = spawn(pythonCommand, [...pythonBaseArgs, '-c', script], {
         cwd: autoBuildSource,
         env: {
           ...process.env,

auto-claude-ui/src/main/terminal/claude-integration-handler.ts

@@ -10,6 +10,8 @@ import { IPC_CHANNELS } from '../../shared/constants';
 import { getClaudeProfileManager } from '../claude-profile-manager';
 import * as OutputParser from './output-parser';
 import * as SessionHandler from './session-handler';
+import { debugLog, debugError } from '../../shared/utils/debug-logger';
+import { escapeShellArg, buildCdCommand } from '../../shared/utils/shell-escape';
 import type {
   TerminalProcess,
   WindowGetter,
@@ -92,9 +94,11 @@ export function handleOAuthToken(
   console.warn('[ClaudeIntegration] OAuth token detected, length:', token.length);
 
   const email = OutputParser.extractEmail(terminal.outputBuffer);
-  const profileIdMatch = terminal.id.match(/claude-login-(profile-\d+)-/);
+  // Match both custom profiles (profile-123456) and the default profile
+  const profileIdMatch = terminal.id.match(/claude-login-(profile-\d+|default)-/);
 
   if (profileIdMatch) {
+    // Save to specific profile (profile login terminal)
     const profileId = profileIdMatch[1];
     const profileManager = getClaudeProfileManager();
     const success = profileManager.setProfileToken(profileId, token, email || undefined);
@@ -116,16 +120,56 @@ export function handleOAuthToken(
       console.error('[ClaudeIntegration] Failed to save OAuth token to profile:', profileId);
     }
   } else {
-    console.warn('[ClaudeIntegration] OAuth token detected but not in a profile login terminal');
-    const win = getWindow();
-    if (win) {
-      win.webContents.send(IPC_CHANNELS.TERMINAL_OAUTH_TOKEN, {
-        terminalId: terminal.id,
-        email,
-        success: false,
-        message: 'Token detected but no profile associated with this terminal',
-        detectedAt: new Date().toISOString()
-      } as OAuthTokenEvent);
+    // No profile-specific terminal, save to active profile (GitHub OAuth flow, etc.)
+    console.warn('[ClaudeIntegration] OAuth token detected in non-profile terminal, saving to active profile');
+    const profileManager = getClaudeProfileManager();
+    const activeProfile = profileManager.getActiveProfile();
+
+    // Defensive null check for active profile
+    if (!activeProfile) {
+      console.error('[ClaudeIntegration] Failed to save OAuth token: no active profile found');
+      const win = getWindow();
+      if (win) {
+        win.webContents.send(IPC_CHANNELS.TERMINAL_OAUTH_TOKEN, {
+          terminalId: terminal.id,
+          profileId: undefined,
+          email,
+          success: false,
+          message: 'No active profile found',
+          detectedAt: new Date().toISOString()
+        } as OAuthTokenEvent);
+      }
+      return;
+    }
+
+    const success = profileManager.setProfileToken(activeProfile.id, token, email || undefined);
+
+    if (success) {
+      console.warn('[ClaudeIntegration] OAuth token auto-saved to active profile:', activeProfile.name);
+
+      const win = getWindow();
+      if (win) {
+        win.webContents.send(IPC_CHANNELS.TERMINAL_OAUTH_TOKEN, {
+          terminalId: terminal.id,
+          profileId: activeProfile.id,
+          email,
+          success: true,
+          detectedAt: new Date().toISOString()
+        } as OAuthTokenEvent);
+      }
+    } else {
+      console.error('[ClaudeIntegration] Failed to save OAuth token to active profile:', activeProfile.name);
+      const win = getWindow();
+      if (win) {
+        win.webContents.send(IPC_CHANNELS.TERMINAL_OAUTH_TOKEN, {
+          terminalId: terminal.id,
+          profileId: activeProfile?.id,
+          email,
+          success: false,
+          message: 'Failed to save token to active profile',
+          detectedAt: new Date().toISOString()
+        } as OAuthTokenEvent);
+      }
     }
   }
 }
@@ -161,6 +205,11 @@ export function invokeClaude(
   getWindow: WindowGetter,
   onSessionCapture: (terminalId: string, projectPath: string, startTime: number) => void
 ): void {
+  debugLog('[ClaudeIntegration:invokeClaude] ========== INVOKE CLAUDE START ==========');
+  debugLog('[ClaudeIntegration:invokeClaude] Terminal ID:', terminal.id);
+  debugLog('[ClaudeIntegration:invokeClaude] Requested profile ID:', profileId);
+  debugLog('[ClaudeIntegration:invokeClaude] CWD:', cwd);
+
   terminal.isClaudeMode = true;
   terminal.claudeSessionId = undefined;
 
@@ -175,31 +224,70 @@ export function invokeClaude(
   const previousProfileId = terminal.claudeProfileId;
   terminal.claudeProfileId = activeProfile?.id;
 
-  const cwdCommand = cwd ? `cd "${cwd}" && ` : '';
+  debugLog('[ClaudeIntegration:invokeClaude] Profile resolution:', {
+    previousProfileId,
+    newProfileId: activeProfile?.id,
+    profileName: activeProfile?.name,
+    hasOAuthToken: !!activeProfile?.oauthToken,
+    isDefault: activeProfile?.isDefault
+  });
+
+  // Use safe shell escaping to prevent command injection
+  const cwdCommand = buildCdCommand(cwd);
   const needsEnvOverride = profileId && profileId !== previousProfileId;
 
+  debugLog('[ClaudeIntegration:invokeClaude] Environment override check:', {
+    profileIdProvided: !!profileId,
+    previousProfileId,
+    needsEnvOverride
+  });
+
   if (needsEnvOverride && activeProfile && !activeProfile.isDefault) {
     const token = profileManager.getProfileToken(activeProfile.id);
+    debugLog('[ClaudeIntegration:invokeClaude] Token retrieval:', {
+      hasToken: !!token,
+      tokenLength: token?.length
+    });
 
     if (token) {
       const tempFile = path.join(os.tmpdir(), `.claude-token-${Date.now()}`);
+      debugLog('[ClaudeIntegration:invokeClaude] Writing token to temp file:', tempFile);
       fs.writeFileSync(tempFile, `export CLAUDE_CODE_OAUTH_TOKEN="${token}"\n`, { mode: 0o600 });
 
-      terminal.pty.write(`${cwdCommand}source "${tempFile}" && rm -f "${tempFile}" && claude\r`);
-      console.warn('[ClaudeIntegration] Switching to Claude profile:', activeProfile.name, '(via secure temp file)');
+      // Clear terminal and run command without adding to shell history:
+      // - HISTFILE= disables history file writing for the current command
+      // - HISTCONTROL=ignorespace causes commands starting with space to be ignored
+      // - Leading space ensures the command is ignored even if HISTCONTROL was already set
+      // - Uses subshell (...) to isolate environment changes
+      // This prevents temp file paths from appearing in shell history
+      const command = `clear && ${cwdCommand} HISTFILE= HISTCONTROL=ignorespace bash -c 'source "${tempFile}" && rm -f "${tempFile}" && exec claude'\r`;
+      debugLog('[ClaudeIntegration:invokeClaude] Executing command (temp file method, history-safe)');
+      terminal.pty.write(command);
+      debugLog('[ClaudeIntegration:invokeClaude] ========== INVOKE CLAUDE COMPLETE (temp file) ==========');
       return;
     } else if (activeProfile.configDir) {
-      terminal.pty.write(`${cwdCommand}CLAUDE_CONFIG_DIR="${activeProfile.configDir}" claude\r`);
-      console.warn('[ClaudeIntegration] Using Claude profile:', activeProfile.name, 'config:', activeProfile.configDir);
+      // Clear terminal and run command without adding to shell history:
+      // Same history-disabling technique as temp file method above
+      // SECURITY: Use escapeShellArg for configDir to prevent command injection
+      // Set CLAUDE_CONFIG_DIR as env var before bash -c to avoid embedding user input in the command string
+      const escapedConfigDir = escapeShellArg(activeProfile.configDir);
+      const command = `clear && ${cwdCommand}HISTFILE= HISTCONTROL=ignorespace CLAUDE_CONFIG_DIR=${escapedConfigDir} bash -c 'exec claude'\r`;
+      debugLog('[ClaudeIntegration:invokeClaude] Executing command (configDir method, history-safe)');
+      terminal.pty.write(command);
+      debugLog('[ClaudeIntegration:invokeClaude] ========== INVOKE CLAUDE COMPLETE (configDir) ==========');
       return;
+    } else {
+      debugLog('[ClaudeIntegration:invokeClaude] WARNING: No token or configDir available for non-default profile');
     }
   }
 
   if (activeProfile && !activeProfile.isDefault) {
-    console.warn('[ClaudeIntegration] Using Claude profile:', activeProfile.name, '(from terminal environment)');
+    debugLog('[ClaudeIntegration:invokeClaude] Using terminal environment for non-default profile:', activeProfile.name);
   }
 
-  terminal.pty.write(`${cwdCommand}claude\r`);
+  const command = `${cwdCommand}claude\r`;
+  debugLog('[ClaudeIntegration:invokeClaude] Executing command (default method):', command);
+  terminal.pty.write(command);
 
   if (activeProfile) {
     profileManager.markProfileUsed(activeProfile.id);
@@ -220,6 +308,8 @@ export function invokeClaude(
   if (projectPath) {
     onSessionCapture(terminal.id, projectPath, startTime);
   }
+
+  debugLog('[ClaudeIntegration:invokeClaude] ========== INVOKE CLAUDE COMPLETE (default) ==========');
 }
 
 /**
@@ -234,7 +324,8 @@ export function resumeClaude(
 
   let command: string;
   if (sessionId) {
-    command = `claude --resume "${sessionId}"`;
+    // SECURITY: Escape sessionId to prevent command injection
+    command = `claude --resume ${escapeShellArg(sessionId)}`;
     terminal.claudeSessionId = sessionId;
   } else {
     command = 'claude --continue';
@@ -248,6 +339,103 @@ export function resumeClaude(
   }
 }
 
+/**
+ * Configuration for waiting for Claude to exit
+ */
+interface WaitForExitConfig {
+  /** Maximum time to wait for Claude to exit (ms) */
+  timeout?: number;
+  /** Interval between checks (ms) */
+  pollInterval?: number;
+}
+
+/**
+ * Result of waiting for Claude to exit
+ */
+interface WaitForExitResult {
+  /** Whether Claude exited successfully */
+  success: boolean;
+  /** Error message if failed */
+  error?: string;
+  /** Whether the operation timed out */
+  timedOut?: boolean;
+}
+
+/**
+ * Shell prompt patterns that indicate Claude has exited and shell is ready
+ * These patterns match common shell prompts across bash, zsh, fish, etc.
+ */
+const SHELL_PROMPT_PATTERNS = [
+  /[$%#>❯]\s*$/m,                    // Common prompt endings: $, %, #, >, ❯
+  /\w+@[\w.-]+[:\s]/,                // user@hostname: format
+  /^\s*\S+\s*[$%#>❯]\s*$/m,          // hostname/path followed by prompt char
+  /\(.*\)\s*[$%#>❯]\s*$/m,           // (venv) or (branch) followed by prompt
+];
+
+/**
+ * Wait for Claude to exit by monitoring terminal output for shell prompt
+ *
+ * Instead of using fixed delays, this monitors the terminal's outputBuffer
+ * for patterns indicating that Claude has exited and the shell prompt is visible.
+ */
+async function waitForClaudeExit(
+  terminal: TerminalProcess,
+  config: WaitForExitConfig = {}
+): Promise<WaitForExitResult> {
+  const { timeout = 5000, pollInterval = 100 } = config;
+
+  debugLog('[ClaudeIntegration:waitForClaudeExit] Waiting for Claude to exit...');
+  debugLog('[ClaudeIntegration:waitForClaudeExit] Config:', { timeout, pollInterval });
+
+  // Capture current buffer length to detect new output
+  const initialBufferLength = terminal.outputBuffer.length;
+  const startTime = Date.now();
+
+  return new Promise((resolve) => {
+    const checkForPrompt = () => {
+      const elapsed = Date.now() - startTime;
+
+      // Check for timeout
+      if (elapsed >= timeout) {
+        console.warn('[ClaudeIntegration:waitForClaudeExit] Timeout waiting for Claude to exit after', timeout, 'ms');
+        debugLog('[ClaudeIntegration:waitForClaudeExit] Timeout reached, Claude may not have exited cleanly');
+        resolve({
+          success: false,
+          error: `Timeout waiting for Claude to exit after ${timeout}ms`,
+          timedOut: true
+        });
+        return;
+      }
+
+      // Get new output since we started waiting
+      const newOutput = terminal.outputBuffer.slice(initialBufferLength);
+
+      // Check if we can see a shell prompt in the new output
+      for (const pattern of SHELL_PROMPT_PATTERNS) {
+        if (pattern.test(newOutput)) {
+          debugLog('[ClaudeIntegration:waitForClaudeExit] Shell prompt detected after', elapsed, 'ms');
+          debugLog('[ClaudeIntegration:waitForClaudeExit] Matched pattern:', pattern.toString());
+          resolve({ success: true });
+          return;
+        }
+      }
+
+      // Also check if isClaudeMode was cleared (set by other handlers)
+      if (!terminal.isClaudeMode) {
+        debugLog('[ClaudeIntegration:waitForClaudeExit] isClaudeMode flag cleared after', elapsed, 'ms');
+        resolve({ success: true });
+        return;
+      }
+
+      // Continue polling
+      setTimeout(checkForPrompt, pollInterval);
+    };
+
+    // Start checking
+    checkForPrompt();
+  });
+}
+
 /**
  * Switch terminal to a different Claude profile
  */
@@ -258,27 +446,95 @@ export async function switchClaudeProfile(
   invokeClaudeCallback: (terminalId: string, cwd: string | undefined, profileId: string) => void,
   clearRateLimitCallback: (terminalId: string) => void
 ): Promise<{ success: boolean; error?: string }> {
+  // Always-on tracing
+  console.warn('[ClaudeIntegration:switchClaudeProfile] Called for terminal:', terminal.id, '| profileId:', profileId);
+  console.warn('[ClaudeIntegration:switchClaudeProfile] Terminal state: isClaudeMode=', terminal.isClaudeMode);
+
+  debugLog('[ClaudeIntegration:switchClaudeProfile] ========== SWITCH PROFILE START ==========');
+  debugLog('[ClaudeIntegration:switchClaudeProfile] Terminal ID:', terminal.id);
+  debugLog('[ClaudeIntegration:switchClaudeProfile] Target profile ID:', profileId);
+  debugLog('[ClaudeIntegration:switchClaudeProfile] Terminal state:', {
+    isClaudeMode: terminal.isClaudeMode,
+    currentProfileId: terminal.claudeProfileId,
+    claudeSessionId: terminal.claudeSessionId,
+    projectPath: terminal.projectPath,
+    cwd: terminal.cwd
+  });
+
   const profileManager = getClaudeProfileManager();
   const profile = profileManager.getProfile(profileId);
+
+  console.warn('[ClaudeIntegration:switchClaudeProfile] Profile found:', profile?.name || 'NOT FOUND');
+  debugLog('[ClaudeIntegration:switchClaudeProfile] Target profile:', profile ? {
+    id: profile.id,
+    name: profile.name,
+    hasOAuthToken: !!profile.oauthToken,
+    isDefault: profile.isDefault
+  } : 'NOT FOUND');
+
   if (!profile) {
+    console.error('[ClaudeIntegration:switchClaudeProfile] Profile not found, aborting');
+    debugError('[ClaudeIntegration:switchClaudeProfile] Profile not found, aborting');
     return { success: false, error: 'Profile not found' };
   }
 
-  console.warn('[ClaudeIntegration] Switching to Claude profile:', profile.name);
+  console.warn('[ClaudeIntegration:switchClaudeProfile] Switching to profile:', profile.name);
+  debugLog('[ClaudeIntegration:switchClaudeProfile] Switching to Claude profile:', profile.name);
 
   if (terminal.isClaudeMode) {
+    console.warn('[ClaudeIntegration:switchClaudeProfile] Sending exit commands (Ctrl+C, /exit)');
+    debugLog('[ClaudeIntegration:switchClaudeProfile] Terminal is in Claude mode, sending exit commands');
+
+    // Send Ctrl+C to interrupt any ongoing operation
+    debugLog('[ClaudeIntegration:switchClaudeProfile] Sending Ctrl+C (\\x03)');
     terminal.pty.write('\x03');
-    await new Promise(resolve => setTimeout(resolve, 500));
+
+    // Wait briefly for Ctrl+C to take effect before sending /exit
+    await new Promise(resolve => setTimeout(resolve, 100));
+
+    // Send /exit command
+    debugLog('[ClaudeIntegration:switchClaudeProfile] Sending /exit command');
     terminal.pty.write('/exit\r');
-    await new Promise(resolve => setTimeout(resolve, 500));
+
+    // Wait for Claude to actually exit by monitoring for shell prompt
+    const exitResult = await waitForClaudeExit(terminal, { timeout: 5000, pollInterval: 100 });
+
+    if (exitResult.timedOut) {
+      console.warn('[ClaudeIntegration:switchClaudeProfile] Timed out waiting for Claude to exit, proceeding with caution');
+      debugLog('[ClaudeIntegration:switchClaudeProfile] Exit timeout - terminal may be in inconsistent state');
+
+      // Even on timeout, we'll try to proceed but log the warning
+      // The alternative would be to abort, but that could leave users stuck
+      // If this becomes a problem, we could add retry logic or abort option
+    } else if (!exitResult.success) {
+      console.error('[ClaudeIntegration:switchClaudeProfile] Failed to exit Claude:', exitResult.error);
+      debugError('[ClaudeIntegration:switchClaudeProfile] Exit failed:', exitResult.error);
+      // Continue anyway - the /exit command was sent
+    } else {
+      console.warn('[ClaudeIntegration:switchClaudeProfile] Claude exited successfully');
+      debugLog('[ClaudeIntegration:switchClaudeProfile] Claude exited, ready to switch profile');
+    }
+  } else {
+    console.warn('[ClaudeIntegration:switchClaudeProfile] NOT in Claude mode, skipping exit commands');
+    debugLog('[ClaudeIntegration:switchClaudeProfile] Terminal NOT in Claude mode, skipping exit commands');
   }
 
+  debugLog('[ClaudeIntegration:switchClaudeProfile] Clearing rate limit state for terminal');
   clearRateLimitCallback(terminal.id);
 
   const projectPath = terminal.projectPath || terminal.cwd;
+  console.warn('[ClaudeIntegration:switchClaudeProfile] Invoking Claude with profile:', profileId, '| cwd:', projectPath);
+  debugLog('[ClaudeIntegration:switchClaudeProfile] Invoking Claude with new profile:', {
+    terminalId: terminal.id,
+    projectPath,
+    profileId
+  });
   invokeClaudeCallback(terminal.id, projectPath, profileId);
 
+  debugLog('[ClaudeIntegration:switchClaudeProfile] Setting active profile in profile manager');
   profileManager.setActiveProfile(profileId);
 
+  console.warn('[ClaudeIntegration:switchClaudeProfile] COMPLETE');
+  debugLog('[ClaudeIntegration:switchClaudeProfile] ========== SWITCH PROFILE COMPLETE ==========');
   return { success: true };
 }

auto-claude-ui/src/main/terminal/pty-daemon.ts

@@ -10,7 +10,7 @@
 
 import * as net from 'net';
 import * as fs from 'fs';
-import * as pty from 'node-pty';
+import * as pty from '@lydell/node-pty';
 
 const SOCKET_PATH =
   process.platform === 'win32'

auto-claude-ui/src/main/terminal/pty-manager.ts

@@ -3,7 +3,7 @@
  * Handles low-level PTY process creation and lifecycle
  */
 
-import * as pty from 'node-pty';
+import * as pty from '@lydell/node-pty';
 import * as os from 'os';
 import type { TerminalProcess, WindowGetter } from './types';
 import { IPC_CHANNELS } from '../../shared/constants';

auto-claude-ui/src/main/terminal/types.ts

@@ -1,4 +1,4 @@
-import type * as pty from 'node-pty';
+import type * as pty from '@lydell/node-pty';
 import type { BrowserWindow } from 'electron';
 
 /**

auto-claude-ui/src/main/title-generator.ts

@@ -4,11 +4,12 @@ import { spawn } from 'child_process';
 import { app } from 'electron';
 import { EventEmitter } from 'events';
 import { detectRateLimit, createSDKRateLimitInfo, getProfileEnv } from './rate-limit-detector';
+import { findPythonCommand, parsePythonCommand } from './python-detector';
 
 /**
- * Debug logging - only logs when AUTO_CLAUDE_DEBUG env var is set
+ * Debug logging - only logs when DEBUG=true or in development mode
  */
-const DEBUG = process.env.AUTO_CLAUDE_DEBUG === 'true' || process.env.AUTO_CLAUDE_DEBUG === '1';
+const DEBUG = process.env.DEBUG === 'true' || process.env.NODE_ENV === 'development';
 
 function debug(...args: unknown[]): void {
   if (DEBUG) {
@@ -20,7 +21,8 @@ function debug(...args: unknown[]): void {
  * Service for generating task titles from descriptions using Claude AI
  */
 export class TitleGenerator extends EventEmitter {
-  private pythonPath: string = 'python3';
+  // Auto-detect Python command on initialization
+  private pythonPath: string = findPythonCommand() || 'python';
   private autoBuildSourcePath: string = '';
 
   constructor() {
@@ -129,7 +131,9 @@ export class TitleGenerator extends EventEmitter {
     const profileEnv = getProfileEnv();
 
     return new Promise((resolve) => {
-      const childProcess = spawn(this.pythonPath, ['-c', script], {
+      // Parse Python command to handle space-separated commands like "py -3"
+      const [pythonCommand, pythonBaseArgs] = parsePythonCommand(this.pythonPath);
+      const childProcess = spawn(pythonCommand, [...pythonBaseArgs, '-c', script], {
         cwd: autoBuildSource,
         env: {
           ...process.env,

auto-claude-ui/src/main/updater/http-client.ts

@@ -74,9 +74,12 @@ export function downloadFile(
   return new Promise((resolve, reject) => {
     const file = createWriteStream(destPath);
 
+    // GitHub API URLs need the GitHub Accept header to get a redirect to the actual file
+    // Non-API URLs (CDN, direct downloads) use octet-stream
+    const isGitHubApi = url.includes('api.github.com');
     const headers = {
       'User-Agent': 'Auto-Claude-UI',
-      'Accept': 'application/octet-stream'
+      'Accept': isGitHubApi ? 'application/vnd.github+json' : 'application/octet-stream'
     };
 
     const request = https.get(url, { headers }, (response) => {

auto-claude-ui/src/main/updater/update-checker.ts

@@ -4,8 +4,9 @@
 
 import { GITHUB_CONFIG } from './config';
 import { fetchJson } from './http-client';
-import { getBundledVersion, parseVersionFromTag, compareVersions } from './version-manager';
+import { getEffectiveVersion, parseVersionFromTag, compareVersions } from './version-manager';
 import { GitHubRelease, AutoBuildUpdateCheck } from './types';
+import { debugLog } from '../../shared/utils/debug-logger';
 
 // Cache for the latest release info (used by download)
 let cachedLatestRelease: GitHubRelease | null = null;
@@ -35,7 +36,9 @@ export function clearCachedRelease(): void {
  * Check GitHub Releases for the latest version
  */
 export async function checkForUpdates(): Promise<AutoBuildUpdateCheck> {
-  const currentVersion = getBundledVersion();
+  // Use effective version which accounts for source updates
+  const currentVersion = getEffectiveVersion();
+  debugLog('[UpdateCheck] Current effective version:', currentVersion);
 
   try {
     // Fetch latest release from GitHub Releases API
@@ -47,9 +50,11 @@ export async function checkForUpdates(): Promise<AutoBuildUpdateCheck> {
 
     // Parse version from tag (e.g., "v1.2.0" -> "1.2.0")
     const latestVersion = parseVersionFromTag(release.tag_name);
+    debugLog('[UpdateCheck] Latest version:', latestVersion);
 
     // Compare versions
     const updateAvailable = compareVersions(latestVersion, currentVersion) > 0;
+    debugLog('[UpdateCheck] Update available:', updateAvailable);
 
     return {
       updateAvailable,
@@ -61,6 +66,7 @@ export async function checkForUpdates(): Promise<AutoBuildUpdateCheck> {
   } catch (error) {
     // Clear cache on error
     clearCachedRelease();
+    debugLog('[UpdateCheck] Error:', error instanceof Error ? error.message : error);
 
     return {
       updateAvailable: false,

auto-claude-ui/src/main/updater/update-installer.ts

@@ -12,6 +12,7 @@ import { getUpdateCachePath, getUpdateTargetPath } from './path-resolver';
 import { extractTarball, copyDirectoryRecursive, preserveFiles, restoreFiles, cleanTargetDirectory } from './file-operations';
 import { getCachedRelease, setCachedRelease, clearCachedRelease } from './update-checker';
 import { GitHubRelease, AutoBuildUpdateResult, UpdateProgressCallback, UpdateMetadata } from './types';
+import { debugLog } from '../../shared/utils/debug-logger';
 
 /**
  * Download and apply the latest auto-claude update from GitHub Releases
@@ -25,6 +26,9 @@ export async function downloadAndApplyUpdate(
 ): Promise<AutoBuildUpdateResult> {
   const cachePath = getUpdateCachePath();
 
+  debugLog('[Update] Starting update process...');
+  debugLog('[Update] Cache path:', cachePath);
+
   try {
     onProgress?.({
       stage: 'checking',
@@ -34,19 +38,26 @@ export async function downloadAndApplyUpdate(
     // Ensure cache directory exists
     if (!existsSync(cachePath)) {
       mkdirSync(cachePath, { recursive: true });
+      debugLog('[Update] Created cache directory');
     }
 
     // Get release info (use cache or fetch fresh)
     let release = getCachedRelease();
     if (!release) {
       const releaseUrl = `https://api.github.com/repos/${GITHUB_CONFIG.owner}/${GITHUB_CONFIG.repo}/releases/latest`;
+      debugLog('[Update] Fetching release info from:', releaseUrl);
       release = await fetchJson<GitHubRelease>(releaseUrl);
       setCachedRelease(release);
+    } else {
+      debugLog('[Update] Using cached release info');
     }
 
-    // Use the release tarball URL
-    const tarballUrl = release.tarball_url;
+    // Use explicit tag reference URL to avoid HTTP 300 when branch/tag names collide
+    // See: https://github.com/AndyMik90/Auto-Claude/issues/78
+    const tarballUrl = `https://api.github.com/repos/${GITHUB_CONFIG.owner}/${GITHUB_CONFIG.repo}/tarball/refs/tags/${release.tag_name}`;
     const releaseVersion = parseVersionFromTag(release.tag_name);
+    debugLog('[Update] Release version:', releaseVersion);
+    debugLog('[Update] Tarball URL:', tarballUrl);
 
     const tarballPath = path.join(cachePath, 'auto-claude-update.tar.gz');
     const extractPath = path.join(cachePath, 'extracted');
@@ -63,6 +74,8 @@ export async function downloadAndApplyUpdate(
       message: 'Downloading update...'
     });
 
+    debugLog('[Update] Starting download to:', tarballPath);
+
     // Download the tarball
     await downloadFile(tarballUrl, tarballPath, (percent) => {
       onProgress?.({
@@ -72,14 +85,20 @@ export async function downloadAndApplyUpdate(
       });
     });
 
+    debugLog('[Update] Download complete');
+
     onProgress?.({
       stage: 'extracting',
       message: 'Extracting update...'
     });
 
+    debugLog('[Update] Extracting to:', extractPath);
+
     // Extract the tarball
     await extractTarball(tarballPath, extractPath);
 
+    debugLog('[Update] Extraction complete');
+
     // Find the auto-claude folder in extracted content
     // GitHub tarballs have a root folder like "owner-repo-hash/"
     const extractedDirs = readdirSync(extractPath);
@@ -96,6 +115,7 @@ export async function downloadAndApplyUpdate(
 
     // Determine where to install the update
     const targetPath = getUpdateTargetPath();
+    debugLog('[Update] Target install path:', targetPath);
 
     // Backup existing source (if in dev mode)
     const backupPath = path.join(cachePath, 'backup');
@@ -104,11 +124,14 @@ export async function downloadAndApplyUpdate(
         rmSync(backupPath, { recursive: true, force: true });
       }
       // Simple copy for backup
+      debugLog('[Update] Creating backup at:', backupPath);
       copyDirectoryRecursive(targetPath, backupPath);
     }
 
     // Apply the update
+    debugLog('[Update] Applying update...');
     await applyUpdate(targetPath, autoBuildSource);
+    debugLog('[Update] Update applied successfully');
 
     // Write update metadata
     const metadata: UpdateMetadata = {
@@ -132,14 +155,26 @@ export async function downloadAndApplyUpdate(
       message: `Updated to version ${releaseVersion}`
     });
 
+    debugLog('[Update] ============================================');
+    debugLog('[Update] UPDATE SUCCESSFUL');
+    debugLog('[Update] New version:', releaseVersion);
+    debugLog('[Update] Target path:', targetPath);
+    debugLog('[Update] ============================================');
+
     return {
       success: true,
       version: releaseVersion
     };
   } catch (error) {
+    const errorMessage = error instanceof Error ? error.message : 'Update failed';
+    debugLog('[Update] ============================================');
+    debugLog('[Update] UPDATE FAILED');
+    debugLog('[Update] Error:', errorMessage);
+    debugLog('[Update] ============================================');
+
     onProgress?.({
       stage: 'error',
-      message: error instanceof Error ? error.message : 'Update failed'
+      message: errorMessage
     });
 
     return {

auto-claude-ui/src/main/updater/version-manager.ts

@@ -3,17 +3,85 @@
  */
 
 import { app } from 'electron';
+import { existsSync, readFileSync } from 'fs';
+import path from 'path';
+import type { UpdateMetadata } from './types';
 
 /**
- * Get the current app/framework version
+ * Get the current app/framework version from package.json
  *
- * Uses app.getVersion() (from package.json) as the single source of truth.
- * Both the Electron app and auto-claude framework share the same version.
+ * Uses app.getVersion() (from package.json) as the base version.
  */
 export function getBundledVersion(): string {
   return app.getVersion();
 }
 
+/**
+ * Get the effective version - accounts for source updates
+ *
+ * Returns the updated source version if an update has been applied,
+ * otherwise returns the bundled version.
+ */
+export function getEffectiveVersion(): string {
+  const isDebug = process.env.DEBUG === 'true';
+
+  // Build list of paths to check for update metadata
+  const metadataPaths: string[] = [];
+
+  if (app.isPackaged) {
+    // Production: check userData override path
+    metadataPaths.push(
+      path.join(app.getPath('userData'), 'auto-claude-source', '.update-metadata.json')
+    );
+  } else {
+    // Development: check the actual source paths where updates are written
+    const possibleSourcePaths = [
+      path.join(app.getAppPath(), '..', 'auto-claude'),
+      path.join(app.getAppPath(), '..', '..', 'auto-claude'),
+      path.join(process.cwd(), 'auto-claude'),
+      path.join(process.cwd(), '..', 'auto-claude')
+    ];
+
+    for (const sourcePath of possibleSourcePaths) {
+      metadataPaths.push(path.join(sourcePath, '.update-metadata.json'));
+    }
+  }
+
+  if (isDebug) {
+    console.log('[Version] Checking metadata paths:', metadataPaths);
+  }
+
+  // Check each path for metadata
+  for (const metadataPath of metadataPaths) {
+    const exists = existsSync(metadataPath);
+    if (isDebug) {
+      console.log(`[Version] Checking ${metadataPath}: ${exists ? 'EXISTS' : 'not found'}`);
+    }
+    if (exists) {
+      try {
+        const metadata = JSON.parse(readFileSync(metadataPath, 'utf-8')) as UpdateMetadata;
+        if (metadata.version) {
+          if (isDebug) {
+            console.log(`[Version] Found metadata version: ${metadata.version}`);
+          }
+          return metadata.version;
+        }
+      } catch (e) {
+        if (isDebug) {
+          console.log(`[Version] Error reading metadata: ${e}`);
+        }
+        // Continue to next path
+      }
+    }
+  }
+
+  const bundledVersion = app.getVersion();
+  if (isDebug) {
+    console.log(`[Version] No metadata found, using bundled version: ${bundledVersion}`);
+  }
+  return bundledVersion;
+}
+
 /**
  * Parse version from GitHub release tag
  * Handles tags like "v1.2.0", "1.2.0", "v1.2.0-beta"

auto-claude-ui/src/preload/api/modules/changelog-api.ts

@@ -48,6 +48,10 @@ export interface ChangelogAPI {
     imageData: string,
     filename: string
   ) => Promise<IPCResult<{ relativePath: string; url: string }>>;
+  readLocalImage: (
+    projectPath: string,
+    relativePath: string
+  ) => Promise<IPCResult<string>>;
 
   // Event Listeners
   onChangelogGenerationProgress: (
@@ -113,6 +117,12 @@ export const createChangelogAPI = (): ChangelogAPI => ({
   ): Promise<IPCResult<{ relativePath: string; url: string }>> =>
     invokeIpc(IPC_CHANNELS.CHANGELOG_SAVE_IMAGE, projectId, imageData, filename),
 
+  readLocalImage: (
+    projectPath: string,
+    relativePath: string
+  ): Promise<IPCResult<string>> =>
+    invokeIpc(IPC_CHANNELS.CHANGELOG_READ_LOCAL_IMAGE, projectPath, relativePath),
+
   // Event Listeners
   onChangelogGenerationProgress: (
     callback: (projectId: string, progress: ChangelogGenerationProgress) => void

auto-claude-ui/src/preload/api/modules/github-api.ts

@@ -41,9 +41,18 @@ export interface GitHubAPI {
   getGitHubUser: () => Promise<IPCResult<{ username: string; name?: string }>>;
   listGitHubUserRepos: () => Promise<IPCResult<{ repos: Array<{ fullName: string; description: string | null; isPrivate: boolean }> }>>;
 
-  // Repository detection
+  // Repository detection and management
   detectGitHubRepo: (projectPath: string) => Promise<IPCResult<string>>;
   getGitHubBranches: (repo: string, token: string) => Promise<IPCResult<string[]>>;
+  createGitHubRepo: (
+    repoName: string,
+    options: { description?: string; isPrivate?: boolean; projectPath: string; owner?: string }
+  ) => Promise<IPCResult<{ fullName: string; url: string }>>;
+  addGitRemote: (
+    projectPath: string,
+    repoFullName: string
+  ) => Promise<IPCResult<{ remoteUrl: string }>>;
+  listGitHubOrgs: () => Promise<IPCResult<{ orgs: Array<{ login: string; avatarUrl?: string }> }>>;
 
   // Event Listeners
   onGitHubInvestigationProgress: (
@@ -113,13 +122,28 @@ export const createGitHubAPI = (): GitHubAPI => ({
   listGitHubUserRepos: (): Promise<IPCResult<{ repos: Array<{ fullName: string; description: string | null; isPrivate: boolean }> }>> =>
     invokeIpc(IPC_CHANNELS.GITHUB_LIST_USER_REPOS),
 
-  // Repository detection
+  // Repository detection and management
   detectGitHubRepo: (projectPath: string): Promise<IPCResult<string>> =>
     invokeIpc(IPC_CHANNELS.GITHUB_DETECT_REPO, projectPath),
 
   getGitHubBranches: (repo: string, token: string): Promise<IPCResult<string[]>> =>
     invokeIpc(IPC_CHANNELS.GITHUB_GET_BRANCHES, repo, token),
 
+  createGitHubRepo: (
+    repoName: string,
+    options: { description?: string; isPrivate?: boolean; projectPath: string; owner?: string }
+  ): Promise<IPCResult<{ fullName: string; url: string }>> =>
+    invokeIpc(IPC_CHANNELS.GITHUB_CREATE_REPO, repoName, options),
+
+  addGitRemote: (
+    projectPath: string,
+    repoFullName: string
+  ): Promise<IPCResult<{ remoteUrl: string }>> =>
+    invokeIpc(IPC_CHANNELS.GITHUB_ADD_REMOTE, projectPath, repoFullName),
+
+  listGitHubOrgs: (): Promise<IPCResult<{ orgs: Array<{ login: string; avatarUrl?: string }> }>> =>
+    invokeIpc(IPC_CHANNELS.GITHUB_LIST_ORGS),
+
   // Event Listeners
   onGitHubInvestigationProgress: (
     callback: (projectId: string, status: GitHubInvestigationStatus) => void