The Production Framework for
Self-Improving Knowledge Vaults
AIのドリフトとハルシネーションを止め、カオスなAIメモを機械的に検証された自己組織化セカンドブレインへ。頻出する修正は自動的に再利用可能なスキルへ昇格。
🧬 自己組織化ライフサイクル
Rotate through the 4-stage closed loop that powers autonomous vault evolution.
1. Immutable Ingestion (Raw)
Never edited by LLM. Original notes, benchmark runs, papers, and logs are stored as ground truth.
2. Grounded Compilation
AI agent compiles raw sources into concise wiki pages with mandatory Raw: provenance links.
3. Mechanical Invariant Check
check_evidence.py verifies every number, date, and quote against raw bodies. 0-error rule.
4. Self-Organizing GC & Auto-Skillify
Outdated facts are archived (never deleted). Repeated fixes (2+) automatically evolve into reusable skills.
Source of Truth Ingestion
All human thoughts, logs, and external data land in raw/notes/ and remain immutable forever.
Provenance Linking
Agents synthesize insights and attach explicit provenance links. No claim without evidence.
0-Hallucination Evidence Check
Strict mechanical matching checks candidate dates, grouped numbers, and quoted strings.
Autonomous Skill Promotion
When a fix appears twice in log.md, the system proposes promoting it into a permanent skill.
🌐 1-Click Multi-Agent Compatibility
Inject standardized skills into all your favorite AI coding assistants with zero configuration.
Claude Code
Direct integration with ~/.claude/skills for terminal agent pairing.
Cursor & Windsurf
Project-level .cursor/skills/ and rules support for seamless in-editor context.
Gemini & OpenCode
Full support for .gemini/skills and open agent specifications.
🚀 10秒で開始
Initialize your first self-improving knowledge vault right now.