菜单
Use when an existing skill needs updating based on new evidence
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Use when processing conversations to extract reusable knowledge (entities, facts, processes)
Use after completing any task to extract reusable patterns
Use when you've identified a recurring pattern worth abstracting into a skill
Use before starting any task to find relevant skills
Advanced Python development guidance with focus on architecture, design patterns, code quality, type safety, and modern best practices. Use when designing Python systems, refactoring codebases, implementing complex features, reviewing architecture decisions, applying design patterns, or ensuring code follows best practices for maintainability, testability, and performance.
| name | skill-editing |
| description | Use when an existing skill needs updating based on new evidence |
| trigger_pattern | When a skill receives negative feedback or new evidence contradicts it |
| tags | ["meta","editing","evolution"] |
| version | 1 |
| is_protected | true |
| q_value | 0.5 |
| q_update_count | 0 |
NO EDIT WITHOUT EVIDENCE.
Every skill edit must be justified by concrete evidence: feedback signals, new processes, or observed failures.
Skill needs attention:
├── Q-value < 0.3 with 5+ uses?
│ ├── Fixable with content update?
│ │ └── YES → Edit content
│ └── Fundamentally wrong approach?
│ └── YES → Deprecate and create new
├── Q-value < 0.1 with 5+ uses?
│ └── Auto-suspended → Review for deprecation
├── New evidence contradicts skill?
│ ├── Minor correction needed?
│ │ └── Edit without version bump
│ └── Major change to approach?
│ └── Edit with version bump
├── Skill content is stale?
│ └── Update with version bump
└── Skill is protected?
└── Requires force=True and clear justification
skill_edit(name, new_content, version_bump=False, reason="Clarified step 3")
skill_edit(name, new_content, version_bump=True, reason="Added heap analysis step based on proc_xxx evidence")
Mark as suspended rather than deleting:
skill_edit(name, metadata_updates={"is_suspended": true}, reason="Superseded by skill-xyz")
Meta-skills (skill-discovery, skill-creation, etc.) are protected:
force=True to edit| Action | Minimum Evidence |
|---|---|
| Minor edit | 1 feedback signal or observation |
| Major edit | 2+ feedback signals or new process data |
| Deprecation | 3+ failure signals or clear supersession |
| Protected edit | Documented reasoning + 2+ evidence points |