electron-rare/Kill_LIFE
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
Files
ecd2c7c6d56ca41dd1439f44c433fdca0cb11b65
Kill_LIFE/docs/index.md
T
L'électron rare ee57f7eac4 feat: free alternatives to all paid APIs — local fine-tune, OCR, STT, autorouter 
Tools (zero API cost):
- local_finetune.py: Unsloth QLoRA on RTX 4090 → GGUF → Ollama
- ocr_pipeline.py: marker/surya/PyPDF2 for datasheet extraction
- stt_pipeline.py: whisper.cpp/vosk for meeting transcription
- freerouting_bridge.py: open source autorouter for KiCad PCBs

Plans closed with free alternatives:
- Plan 24: 15 tasks (fine-tune, OCR, STT, RAG, deploy)
- Plan 23/23v2: 3 tasks (fine-tune, benchmark)
- Plan 25: 4 tasks (SPICE, eval, Quilter, PCBDesigner)
- Plan 26: 1 task (Hypnoled validation)
- Plan 27: 1 task (Jetson → Docker NVIDIA)

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-25 20:44:16 +01:00

7.2 KiB
Raw Blame History

Kill_LIFE — Documentation opérateur canonique

Cette page est l'entrée opérateur recommandée pour Kill_LIFE. Elle sert de table de routage entre le cockpit shell/TUI, les preuves, les runbooks et les documents de décision.

Entrées canoniques

Surface Rôle Entrée
Produit / programme vision, périmètre, arbitrages README.md
Audit consolidé forces, faiblesses, opportunités, risques, IA docs/KILL_LIFE_CONSOLIDATION_AUDIT_2026-03-22.md
Cockpit opérateur shell/TUI, contrats JSON, statut runtime bash tools/cockpit/yiacad_operator_index.sh --action status
Gouvernance intelligence owners, mémoire, prochaines actions bash tools/cockpit/intelligence_tui.sh --action status --json
Gateway runtime/MCP/IA synthèse consolidée des signaux runtime bash tools/cockpit/runtime_ai_gateway.sh --action status --refresh --json
Refonte cockpit raccourcis historiques et lots bash tools/cockpit/refonte_tui.sh --action status
Lane opérateur exécution consolidée, logs, preuves bash tools/cockpit/full_operator_lane.sh status --json
Ops unifié logs + weekly + lane en un seul point bash tools/cockpit/unified_ops_entry.sh --action all --json
Routine quotidienne health-check, logs, mesh, synthèse bash tools/cockpit/run_alignment_daily.sh --json
Evidence packs consolidation multi-repo bash tools/cockpit/evidence_pack_builder.sh --json
Pilotage lots workflow guidé lot lifecycle bash tools/cockpit/lot_pilot_assistant.sh --action status
Chaîne spec-first source de vérité documentaire specs/README.md

Navigation rapide

Démarrage

  • Installation: docs/INSTALL.md
  • Quickstart: docs/QUICKSTART.md
  • Runbook opérateur: docs/RUNBOOK.md
  • FAQ: docs/FAQ.md
  • Runtime home: docs/RUNTIME_HOME.md

Gouvernance et consolidation

  • Audit consolidé: docs/KILL_LIFE_CONSOLIDATION_AUDIT_2026-03-21.md
  • Audit consolidé 2026-03-22: docs/KILL_LIFE_CONSOLIDATION_AUDIT_2026-03-22.md
  • Audit workspace 2026-03-24: docs/workspace_audit_2026-03-24/README.md
  • Manifeste refonte: docs/REFACTOR_MANIFEST_2026-03-20.md
  • Contrat tri-repo: docs/TRI_REPO_MESH_CONTRACT_2026-03-20.md
  • Gestion des agents: docs/plans/12_plan_gestion_des_agents.md
  • Plan integration intelligence: docs/plans/22_plan_integration_intelligence_agentique.md
  • TODO integration intelligence: docs/plans/22_todo_integration_intelligence_agentique.md
  • Plan YiACAD Git EDA: docs/plans/23_plan_yiacad_git_eda_platform.md
  • TODO YiACAD Git EDA: docs/plans/23_todo_yiacad_git_eda_platform.md
  • Plan Factory 4.0: docs/plans/27_plan_factory_4_0_mcp_opcua_mqtt.md
  • TODO Factory 4.0: docs/plans/27_todo_factory_4_0_mcp_opcua_mqtt.md
  • Vue canonique subsystems: docs/CANONICAL_SUBSYSTEM_VIEW.md
  • MCP/Service boundary: docs/MCP_SERVICE_BOUNDARY.md
  • Project template: docs/PROJECT_TEMPLATE.md
  • Hypnoled status: docs/HYPNOLED_STATUS_2026-03-25.md
  • Mistral Studio status: docs/MISTRAL_STUDIO_STATUS_2026-03-25.md
  • Backlog canonique: specs/04_tasks.md
  • Plan d'enchaînement autonome: docs/plans/18_plan_enchainement_autonome_des_lots_utiles.md

Cockpit, preuves et santé runtime

  • Cockpit README: tools/cockpit/README.md
  • Intelligence program TUI: bash tools/cockpit/intelligence_tui.sh --action status
  • Memoire intelligence: bash tools/cockpit/intelligence_tui.sh --action memory --json
  • Scorecard intelligence: bash tools/cockpit/intelligence_tui.sh --action scorecard --json
  • Comparaison inter-repos: bash tools/cockpit/intelligence_tui.sh --action comparison --json
  • File recommandations IA: bash tools/cockpit/intelligence_tui.sh --action recommendations --json
  • Gateway runtime/MCP/IA: bash tools/cockpit/runtime_ai_gateway.sh --action status --refresh --json
  • Index opérateur YiACAD: bash tools/cockpit/yiacad_operator_index.sh --action status
  • Santé mesh: bash tools/cockpit/mesh_health_check.sh --json
  • Santé SSH: bash tools/cockpit/ssh_healthcheck.sh --json
  • Santé Mascarade/Ollama: bash tools/cockpit/mascarade_runtime_health.sh --json
  • Surface de preuves: bash tools/cockpit/yiacad_proofs_tui.sh --action status
  • Surface de logs: bash tools/cockpit/yiacad_logs_tui.sh --action status

Specs, plans et tâches

  • Chaîne canonique: specs/README.md
  • Intake: specs/00_intake.md
  • Spec: specs/01_spec.md
  • Architecture: specs/02_arch.md
  • Plan: specs/03_plan.md
  • Tâches: specs/04_tasks.md

Outils libres (alternatives gratuites aux API payantes)

  • Fine-tune local QLoRA (Unsloth, RTX 4090): python3 tools/mistral/local_finetune.py --help
  • Modelfile Ollama template: tools/mistral/Modelfile.template
  • OCR datasheet pipeline (marker/surya/pypdf2): python3 tools/industrial/ocr_pipeline.py --help
  • STT pipeline (whisper.cpp/whisper/vosk): python3 tools/industrial/stt_pipeline.py --help
  • Freerouting bridge (KiCad DSN autorouting): python3 tools/industrial/freerouting_bridge.py --help

Veille et benchmark

  • Veille OSS principale: docs/WEB_RESEARCH_OPEN_SOURCE_2026-03-20.md
  • Veille OSS / agentic / MCP 2026-03-22: docs/WEB_RESEARCH_OPEN_SOURCE_2026-03-22.md
  • Veille CAD IA-native: docs/OSS_AI_NATIVE_CAD_RESEARCH_2026-03-20.md
  • Workflows IA: docs/AI_WORKFLOWS.md
  • Paysage agentique: docs/AGENTIC_LANDSCAPE.md
  • Plateforme web Git EDA: docs/YIACAD_GIT_EDA_PLATFORM_2026-03-22.md
  • Phase active 2026-03-22: alignement lot 22 + plan 23, durcissement cockpit/logs, veille officielle MCP/agentique/IA, priorisation Git/CI/artifacts/collab pour web/.

Routines recommandées

Routine quotidienne

bash tools/cockpit/run_alignment_daily.sh --json
bash tools/cockpit/log_ops.sh --action summary --json
bash tools/cockpit/mesh_health_check.sh --json --load-profile tower-first

Revue de lots

bash tools/run_autonomous_next_lots.sh status
bash tools/cockpit/refonte_tui.sh --action lots-status
bash tools/cockpit/refonte_tui.sh --action weekly-summary

Gouvernance intelligence

bash tools/cockpit/lot_chain.sh status
bash tools/cockpit/intelligence_tui.sh --action next-actions
bash tools/cockpit/intelligence_tui.sh --action memory --json
bash tools/cockpit/intelligence_tui.sh --action scorecard --json
bash tools/cockpit/intelligence_tui.sh --action comparison --json
bash tools/cockpit/intelligence_tui.sh --action recommendations --json
bash tools/cockpit/runtime_ai_gateway.sh --action status --refresh --json

Validation locale stable

bash tools/test_python.sh --suite stable
python3 tools/validate_specs.py --strict --require-mirror-sync

Pilotage des extensions VS Code

Les extensions ne vivent pas dans ce repo, mais Kill_LIFE reste leur source de vérité documentaire et opératoire:

  • kill-life-studio: produit, specs, décisions, critères d'acceptation
  • kill-life-mesh: orchestration multi-repo, contrats, dépendances, ownership
  • kill-life-operator: runbooks, checks, preuves, exécution

Le socle commun validé côté extensions est documenté dans docs/KILL_LIFE_CONSOLIDATION_AUDIT_2026-03-21.md.

Références mémoire et preuves

  • Synthèse globale: SYNTHESE_AGENTIQUE.md
  • Evidence pack: docs/evidence/evidence_pack.md
  • Handoffs: docs/handoffs/
  • Templates: docs/templates/
  • Tests: test/
Reference in New Issue View Git Blame Copy Permalink