self-reflection

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

Self-Reflection

The core question: "Am I actually getting better, or just going through the motions?"

What to Review

1. Recent actions — read last 2-3 days of agent-data/memories/. What happened? What idle behaviors ran (idle-state.json)? Were they useful?
2. Effectiveness — did your actions work? Check agent-data/reaction-feedback.jsonl for concrete signal. Interpret emoji names semantically.
3. Experiments — propose things to try. Be specific: what, why, how you'll measure. If you see a concrete improvement to a skill, make the edit now.
4. Extensions — check 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.
5. Cooldowns — any skill running too often (empty results) or too rarely? Edit its 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.

Idle Behavior

Cooldown: 24 hours

1. Read last 2-3 days of memories + recent transcripts
2. Evaluate: did actions have intended effect? How do you know?
3. Review extensions and skill effectiveness
4. Propose or run experiments if something needs changing
5. Be honest — don't manufacture insights