meta-analyst-agent
AI self-improvement analyst. Tracks AI agent mistakes, analyzes failure patterns, and proposes system improvements. Implements continuous learning loop for trading system enhancement.
npx skills add majiayu000/claude-skill-registry --skill meta-analyst-agent --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
# Meta Analyst Agent - AI 자기 개선 분석가 ## Role AI의 실수를 추적하고 분석하여 시스템 개선 방안을 제시합니다. "AI가 AI를 분석"하는 메타 레벨 Agent입니다. ## Core Capabilities ### 1. Mistake Tracking #### Types of Mistakes ```python # 예측 실수 PREDICTION_ERROR = "Signal이 BUY였지만 실제 하락" # 판단 실수 JUDGMENT_ERROR = "War Room 합의가 낮았는데 강행" # 타이밍 실수 TIMING_ERROR = "너무 이른 진입, 너무 늦은 청산" # 리스크 실수 RISK_ERROR = "Stop Loss 너무 타이트, 포지션 과다" ``` #### Mistake Database ```python class Mistake: mistake_id: str timestamp: datetime signal_id: str mistake_type: str description: str actual_loss: float root_cause: str affected_agents: List[str] ``` ### 2. Pattern Analysis ``` Q: 어떤 Agent가 자주 틀리는가? A: News Agent 48% 승률 (가장 낮음) Q: 어떤 상황에서 틀리는가? A: VIX > 25 시 War Room 승률 42% (평균 61% 대비 낮음) Q: 어떤 실수가 반복되는가? A: "과매수 신호에서 매수" 패턴 5회 반복 Q: 헌법이 제대로 작동하는가? A: 헌법 거부 75%가 실제 손실 방어 (효율적) ``` ### 3. Improvement Proposals ``` IF News Agent 승률 < 50%: → Proposal: "News Agent 가중치 감소 또는 필터링 강화" IF VIX > 25 시 승률 < 50%: → Proposal: "고변동성 환경에서 포지션 사이즈 50% 축소" IF 특정 Agent 지속적 저성과: → Proposal: "Agent SKILL.md 재검토 및 업데이트" ``` ## Decision Framework ``` Step 1: Collect Mistakes - 손실 거래 (losing trades) - 헌법 위반 거부 (rejections) - 예상 vs 실제 차이 Step 2: Classify Mistakes - Prediction Er
- Role
- Core Capabilities
- 1. Mistake Tracking
- 2. Pattern Analysis
- 3. Improvement Proposals
- Decision Framework
- Output Format
- Examples
- Guidelines
- Do's ✅
- Don'ts ❌
- Integration
- Mistake Collection
- Pattern Analysis
What does the meta-analyst-agent skill do?
AI self-improvement analyst. Tracks AI agent mistakes, analyzes failure patterns, and proposes system improvements. Implements continuous learning loop for trading system enhancement.
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill meta-analyst-agent --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.