pi-skill-system-creator
v0.1.0
Published
A pi skill that creates, extends, and audits skill chain systems — multi-phase skill workflows that pass state through a shared file so they survive across sessions.
Readme

pi-skill-system-creator
A pi coding agent skill that creates, extends, and audits skill chains — multi-phase workflows where each phase is its own skill and state flows between them through a shared file on disk, so the workflow survives across separate sessions and context windows.
This is not a general skill creator (use skill-creator for single-file skills). It specializes in multi-skill chains: designing a new chain, adding a phase to an existing one, fixing broken YAML front matter, and validating that a chain's handoffs and state links are complete.
Install
npm i -g @earendil-works/pi-coding-agent # pi itself, if you don't have it
pi install pi-skill-system-creator # this skillOr straight from git:
pi install git:github.com/KcAnom/pi-skill-system-creatorUse
Once installed, trigger it inside pi with a natural request, e.g.:
- "Create a skill chain that scans a repo, plans fixes, then writes a report."
- "Add a review phase to my
releasechain." - "Audit my chains for cycles, dead ends, and broken handoffs."
- "Fix the YAML front matter in these skills."
Trigger phrases include: create a skill chain, new chain system, add a phase to *, chain audit, validate the chain.
What's in the skill
| File | Purpose |
|---|---|
| skills/skill-system-creator/SKILL.md | The skill itself — rules, templates, and procedures for building and auditing chains. |
| skills/skill-system-creator/EXAMPLE.md | A complete worked example: a real 3-phase demo-scan → demo-plan → demo-report chain with every state transition filled in. |
| skills/skill-system-creator/validate-chain.py | A standalone validator — YAML checks, handoff-graph walk, and cycle / orphan / dead-end detection. Run it instead of improvising audit code. |
What a skill chain is
A skill chain is a group of skills where each skill is one phase of a larger workflow and state flows between phases via a shared state file on disk. Because the state lives on disk rather than in context, the chain can span separate skill invocations — separate sessions, separate context windows — and still pick up exactly where it left off.
License
MIT © KcAnom. Built for the pi coding agent.
