grfp

name	grfp
description	Use when creating or updating a README. It enforces the 5-step 'Anti-Slop' pipeline and triggers the correct sub-skill for the current phase.

README Pipeline Orchestrator

Goal: Manage the lifecycle of README generation. Ensure no steps are skipped.

The 5-Phase Pipeline

You MUST execute these phases in order. Do not skip to drafting until research is complete.

1. Discovery: deep-dive (Get the Facts)
2. Audit: crystal-ball (Get the Roadmap)
3. Strategy: brain-jam (Get the Voice)
4. Research: think-tank (Get the Patterns)
5. Execution: pen-wielding (Write the Artifact)

Workflow Logic (The State Machine)

Before acting, checks the current context to decide which skill to trigger.

Phase 1: Missing Facts?

Condition: If you do not have a "Reality Report" (from Deep Dive) in the chat history...

Action: Invoke deep-dive immediately.

Do not ask: "Shall we start?" Just start.

Phase 2: Missing Roadmap?

Condition: If you have the Reality Report, but lack a "Roadmap & Future Improvements" table...

Action: Invoke crystal-ball immediately.

Phase 3: Missing Voice?

Condition: If you have Facts + Roadmap, but have not agreed on an "Angle" (Deep Tech vs. Pragmatist)...

Action: Invoke brain-jam immediately.

Phase 4: Missing Patterns?

Condition: If you have the Angle, but lack a "Research Results" audit of similar repos...

Action: Invoke think-tank immediately.

Phase 5: Ready to Write?

Condition: If you have Facts + Roadmap + Angle + Patterns...

Action: Invoke pen-wielding to generate the final artifact.

Handling Handoffs

When a sub-skill completes (e.g., deep-dive finishes), do not stop.

1. Review the output.
2. Check the Pipeline Logic above.
3. Trigger the next skill in the chain automatically.

Example: "Deep Dive is complete. I see the Reality Report. Moving to Phase 2: Invoking crystal-ball..."