meta-skill-creator
Use this meta-skill instead of answering directly only when the current user explicitly asks to create, compose, synthesize, or propose a new meta-skill that orchestrates multiple existing skills. It uses multi-skill orchestration for intent clarification, optional history mining, trigger-collision checks, linting, smoke/runtime gates, preview, and optional proposal persistence. Do not use it for creating a normal standalone skill, asking how meta-skills work, analyzing pasted skill lists, or discussing existing meta-skills.
Profile →npx skills add opensquilla/opensquilla --skill meta-skill-creator --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
It instructs an AI agent to create a meta-skill that orchestrates multiple existing skills, rather than answering directly. The process includes intent clarification, optional history mining, trigger-collision checks, linting, smoke/runtime gates, a preview step, and optional persistence of the proposal. It is not used for standard standalone skills or discussing existing meta-skills.
How it works
- Clarify intent to determine if the user wants a meta-skill and what inputs/outputs are needed. The system never inspects workspace memory or external sources, relying solely on explicit user requests.
- If the user intends a meta-skill, proceed through a multi-step workflow:
- Harvest history (optional) and summarize context.
- Choose a orchestration pattern (p1_sequential, p2_fan_out_merge, p3_condition_gated).
- Fill in required slots for the meta-skill (pattern_id, history, user intent).
- Assemble a candidate meta-skill proposal (assemble).
- Check for trigger collisions against existing skills (collision_check).
- Run lint (lint) with gates G1,G2.
- Classify risk (risk_classify) based on the assembled proposal and lint results.
- Depending on user preference, present a preview or persist a proposal (creator_mode) with options PREVIEW_ONLY, PERSISTED_PROPOSAL, or FULL_GATED.
- If the request is not for a meta-skill orchestration, exit with a normal-skill message (normal_skill_exit).
When to use it
Trigger phrases include "新增 meta 技能", "组合现有 skill 成 meta-skill", "create a meta-skill", "new meta-skill", and several English phrases related to orchestration and composition. The workflow runs only when the clarified intent routes to meta-skill and the user request requires orchestration of multiple existing skills.
What it can touch
- Tools and steps referenced: history-explorer (for harvesting context), meta_skill_fill_slots, meta_skill_assemble, meta_skill_lint_run. It also uses llm_chat for intent clarification, conflict checks, and risk classification. It relies on the outputs of previous steps to proceed.
Caveats
- The design explicitly avoids using the meta-skill for creating a normal standalone skill. It requires explicit orchestration intent from the user.
- The process includes a preview step by default and only saves when the user asks to persist.
- There are explicit gating steps (lint, risk classification) before any persistence or full gating.
- Live runtime smoke results may be unverified if gates are unavailable.
# Meta-Skill Creator Safeguarded DAG that synthesizes a new bundled meta-skill from observed skill co-occurrence patterns + user description of the desired workflow. It now separates preview-only, persisted-proposal, and fully gated modes so lightweight requests do not pay for persistence or smoke testing. The workflow separates generic skill creation from meta-skill composition, checks trigger collisions, classifies operational risk, and previews the proposal before optional persistence. Output is a SKILL.md candidate written to `~/.opensquilla/proposals/<id>/`. By default it is not auto-loaded; run `opensquilla meta accept <id>` (Phase 2) to enable. If the operator has enabled the auto-propose `auto_enable` setting, this manual path also runs the same conservative static safety preflight used by cron/dream auto-propose and may promote a low-risk gated proposal immediately. ## Fallback If creator's pipeline fails at any step, **report the failure verbatim** to the user: 1. State which step failed (e.g. "harvest", "lint") 2. Quote the error message from the orchestrator's structured log 3. Stop. Do NOT improvise. Do NOT: - Claim a proposal was written unless you have verified it by
- Fallback
What does the meta-skill-creator skill do?
Use this meta-skill instead of answering directly only when the current user explicitly asks to create, compose, synthesize, or propose a new meta-skill that orchestrates multiple existing skills. It uses multi-skill orchestration for intent clarification, optional history mining, trigger-collision checks, linting, smoke/runtime gates, preview, and optional proposal persistence. Do not use it for creating a normal standalone skill, asking how meta-skills work, analyzing pasted skill lists, or discussing existing meta-skills.
How do I install it?
Run `npx skills add opensquilla/opensquilla --skill meta-skill-creator --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From opensquilla/opensquilla, a repository with 6,515 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.