Menu
Use at the end of a work session to embed learnings into conventions. Simplified for models with lower reasoning capacity.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Use when a file or module is too large, responsibilities are tangled, or a cross-cutting redesign is needed
Use when literature notes exist and need citation verification, or before publishing any artifact that cites literature
Use at the end of a work session, or when accumulated findings need to be embedded into conventions, skills, or patterns
Use when the Slack bot needs to handle an operational action like launching an experiment, querying status, or processing approvals
Use when a plan, finding, or design needs adversarial review before committing to it
Use when designing a new experiment or research protocol and need to ensure methodological rigor
| name | compound-simple |
| description | Use at the end of a work session to embed learnings into conventions. Simplified for models with lower reasoning capacity. |
| complexity | medium |
| model-minimum | fast-model |
Simplified compound skill for Fast Model and similar models. Embed session learnings in <3 turns.
At the end of every autonomous work session, after completing work but before commit.
Execute these steps in order:
Run git diff --stat HEAD~N..HEAD where N is commits this session. Answer:
Answer each question. If yes, note what to update:
Non-obvious fact discovered? (API quirk, hidden constraint, unclear convention) → Update relevant file (CLAUDE.md, skill, or project file)
Failure mode others should avoid? (silent error, misleading config, common mistake) → Add warning/gotcha to relevant skill
Technique that worked well? (debugging strategy, analysis pattern) → If generalizes: note for future gravity evaluation
Convention friction? (rule didn't fit, schema too rigid) → Update convention or note friction
If you completed an experiment/analysis, scan EXPERIMENT.md Findings for:
For each compound opportunity:
| Type | Action |
|---|---|
| Small update (1-3 lines) | Apply directly to file |
| Larger change | Create task in TASKS.md |
| Recurring pattern | Create task: "Run /gravity on: " |
| Governance change | Write to APPROVAL_QUEUE.md |
Output in this format:
Compound: N actions
- <action 1>
- <action 2>
...
If no actions: Compound: no actions this session.
CLAUDE.md — conventions, gotchas.claude/skills/*/SKILL.md — skill instructionsprojects/*/README.md — project-specific learningsprojects/*/TASKS.md — implied follow-up tasks