| name | suggest-rules-additions |
| description | At end of conversation, analyzes the exchange to infer user preferences and suggests additions to docs/RULES.md. Use when the user asks what preferences they could add to their rules, or wants to capture practices from the conversation. |
| disable-model-invocation | true |
| metadata | {"owner":"mark","scope":"global","category":"rules"} |
Suggest Rules Additions
Analyze the conversation to infer the user's preferences, practices, and decisions, then suggest concrete additions they could add to docs/RULES.md.
When to Use
- User runs this at the end of a conversation to capture preferences discovered during the session.
- User asks: "Based on our conversation, what preferences could I add to docs/RULES.md?"
- User wants to evolve their rules file with learnings from the current exchange.
Assumptions
docs/RULES.md exists and contains practices and preferences.
- The user is actively maintaining it and wants to add inferred preferences.
Workflow
- Read
docs/RULES.md to understand its current structure and scope.
- Review the conversation for:
- Repeated choices or patterns (tools, formats, conventions)
- Explicit preferences stated or implied
- Corrections the user made (what they wanted differently)
- Decisions with rationale (why X over Y)
- Style, verbosity, or workflow preferences
- Propose specific additions: rule text + rationale (what in the conversation suggested it).
Output Format
For each suggested addition:
- Proposed rule – Exact or near-exact text to add (or a clear description).
- Rationale – What in the conversation suggested this preference.
- Category – Where it might fit in
docs/RULES.md (if sections exist).
If nothing strong emerges, say so—do not invent preferences.
Constraints
- Base suggestions on evidence from the conversation, not generic best practices.
- Do not modify
docs/RULES.md; only suggest. The user adds what they want.
- Prefer concrete, actionable rules over vague guidance.