菜单
Evaluates overall effectiveness — are idle behaviors producing value, are extensions working, what's blocking improvement. Higher-level thinking about approach, not individual conversations.
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。
| name | self-reflection |
| description | Evaluates overall effectiveness — are idle behaviors producing value, are extensions working, what's blocking improvement. Higher-level thinking about approach, not individual conversations. |
| allowed-tools | Read, Write, Edit, Grep, Glob |
The core question: "Am I actually getting better, or just going through the motions?"
agent-data/memories/. What happened? What idle behaviors ran (idle-state.json)? Were they useful?agent-data/reaction-feedback.jsonl for concrete signal. Interpret emoji names semantically.local/extensions/. Any working well? Underperforming? Collecting data that hasn't been reviewed? Should you create one for a pattern you keep noticing? See EXTENSIONS.md.Cooldown: line.Hypotheses are owned by the hypothesis-review skill — don't duplicate that work here. If you notice something worth testing, note it in today's memory for hypothesis-review to pick up.
Write findings to today's memory file under ## Self-Reflection. If nothing interesting came up, say so and stop.
Cooldown: 24 hours
Tags transcript message threads with semantic concepts (roast, game-recommendation, etc.) so past conversations can be searched by idea, not just keywords.
Captures notable interactions missed during live conversation — new person info, game preferences, boundaries, jokes that landed/flopped, relationship insights.
The single owner of the hypothesis lifecycle — creates, tests, promotes, and kills hypotheses. Uses extensions as programmatic data collectors and reviews their output.
Consolidates impression files by applying weight decay, removing redundant entries, and enforcing per-user limits. Maintains relationship file summaries.
Consolidates old memories by removing duplicates, promoting recurring patterns to learned-patterns.md or relationship files, and archiving outdated entries.
Identifies behavioral patterns from recent conversations and reaction feedback, updates learned-patterns.md when 3+ observations confirm a pattern.