메뉴
Pre-implementation validation requiring ≥90% confidence before coding. Prevents wrong-direction work by assessing duplicates, architecture alignment, and root cause understanding.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Self-improving agent architecture using ChromaDB for continuous learning, self-evaluation, and improvement storage. Agents maintain separate memory collections for learned patterns, performance metrics, and self-assessments without modifying their static .md configuration.
Guide for manually creating custom skills from scratch with templates and validation. Use when designing NEW skills with specific requirements. For auto-generating skills from existing git repos, use /generate-skills instead.
Display QAVR (Q-Value Augmented Vector Retrieval) system status including mode, interaction counts, top memories by Q-value, and configuration.
Enables continuous self-improvement through learning from failures, user corrections, and capability gaps. Integrates with QAVR for learned memory ranking.
Stress-test solutions using the STRIKE framework. Systematically attack proposals to find weaknesses before deployment.
Exhaustive solution space exploration. Use when you need ALL viable options (not just the best), solution space is unknown, or you can't afford to miss alternatives.
| name | confidence-check |
| description | Pre-implementation validation requiring ≥90% confidence before coding. Prevents wrong-direction work by assessing duplicates, architecture alignment, and root cause understanding. |
| metadata | {"clawdbot":{"requires":{"env":[]}}} |
| user-invocable | true |
Pre-implementation validation framework. Never code at < 90% confidence.
Wrong-direction work costs 5K-50K tokens to undo. A 2-minute confidence check saves hours of rework.
Before implementing ANY significant change:
Confidence < 70% → STOP. Research more.
Confidence 70-89% → PAUSE. Fill gaps first.
Confidence ≥ 90% → PROCEED with implementation.
Score each category 0-20:
- 0-5: No understanding
- 6-10: Partial understanding
- 11-15: Good understanding, some gaps
- 16-20: Full understanding
Total = sum of all categories (0-100)
## Confidence Check: [Task]
### Assessment
| Category | Score | Notes |
|----------|-------|-------|
| Duplicate Check | /20 | |
| Architecture | /20 | |
| Requirements | /20 | |
| Root Cause | /20 | |
| Implementation | /20 | |
| **TOTAL** | **/100** | |
### Gaps Identified
- [ ] [gap 1 - how to fill]
- [ ] [gap 2 - how to fill]
### Verdict
**[PROCEED / RESEARCH MORE / STOP]**
### If RESEARCH MORE:
1. [specific question to answer]
2. [file to read]
3. [person to ask]
For small changes, ask yourself:
All Y → Proceed Any N → Pause and fill the gap
❌ "I'll figure it out as I go" ❌ "This looks similar to something I did before" ❌ "The user asked for it so it must be right" ❌ "I'll just try this and see"
✅ "I understand the problem and solution" ✅ "I've verified no existing solution" ✅ "I know exactly what files to change"