agent-skill-kit-ci-integrator
Wire the agent-skills toolkit into CI — install the GitHub Actions workflow that runs validate/lint/audit/deps on every PR, and install the pre-commit hooks for local-loop parity. Use when the user asks to set up CI for skills, install pre-commit hooks, configure GitHub Actions for SKILL.md validation, or wire the toolkit into a new repo.
- License 许可
- Apache-2.0
- 最近更新
- 2026-08-25
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- agent-skill-kit-ci-integrator
- tier
- org
- license
- Apache-2.0
- description
- Wire the agent-skills toolkit into CI — install the GitHub Actions workflow that runs validate/lint/audit/deps on every PR, and install the pre-commit hooks for local-loop parity. Use when the user asks to set up CI for skills, install pre-commit hooks, configure GitHub Actions for SKILL.md validation, or wire the toolkit into a new repo.
- compatibility
- requires `pre-commit` (Python pkg) for local hooks; GitHub Actions for CI; node>=20 + python>=3.11 for the underlying skills-ref CLI
03 SKILL.md 正文
Agent Skill CI Integrator
Documents the two CI integration points the toolkit ships and explains how to install, inspect, and extend them.
When to use
- "Set up CI for our skills"
- "Install pre-commit for SKILL.md"
- "How do I wire skills-ref into GitHub Actions?"
- "Add a PR gate for skills changes"
- "Audit the agent-skills CI configuration"
GitHub Actions workflow
The canonical workflow lives at .github/workflows/agent-skills-ci.yml. It declares three jobs in a strict dependency order:
mjs-ci— runsmake ciin.agents/skills/agent-skill-kit-commons/packages/mjs/. Emitsartifacts/validate.jsonfrom the Node runtime as a stage artifact.py-ci— runsmake ciin.agents/skills/agent-skill-kit-commons/packages/py/. Verifies polyglot parity by re-running the same gates from the Python runtime.gate(depends on both) — runslint, audit, depsfrom the Node runtime, downloadsmjs-ci's validate artifact, and posts a single rolled-up comment on the PR. The comment is keyed by an HTML marker (<!-- agent-skills-ci-marker -->) so subsequent pushes update the existing comment in place rather than stacking new ones.
Triggers: pull_request filtered on paths: [".agents/skills/**", ".github/workflows/agent-skills-ci.yml"] (the .agents/skills/** glob covers the moved agent-skill-kit-commons/packages/** subtree, so no separate filter entry is needed). Permissions: contents: read, pull-requests: write — the minimum needed to comment.
Action pins are all majors (@v4, @v5, @v7); never @main. The security auditor flags @main refs as A030_HOST_PATH_LEAK-adjacent, so the convention is enforced by the toolkit itself.
Pre-commit hooks
The manifest at .pre-commit-hooks.yaml exposes four hooks, one per CLI verb (validate, lint, audit, deps). All hooks use language: system and shell out to the mjs CLI — staying single-runtime in the inner loop avoids paying the parity overhead on every commit.
The local config at .pre-commit-config.yaml consumes the manifest so the suite runs its own hooks against itself. Local install:
pip install pre-commit
pre-commit install
After install, any git commit touching .agents/skills/** runs validate, lint, audit, deps against the entire skill set. pass_filenames: false is critical — the CLI walks the tree itself; without it, pre-commit would replace the args with the changed file paths and break the recursive scan.
Inspect / extend
To add a new gate (e.g., a custom organisation-specific rule):
- Copy a hook entry in
.pre-commit-hooks.yaml, pointentry:at your new tool, keeppass_filenames: false. - Add the matching step to the
gatejob in the workflow YAML. - Run
bash .agents/skills/agent-skill-kit-commons/scripts/test-workflow.shlocally — it parses the YAML, audits for security findings, and confirms all action refs stay pinned to majors.
Self-test driver
.agents/skills/agent-skill-kit-commons/scripts/test-workflow.sh runs offline sanity checks (YAML parse, audit clean, action-ref pinning) and optionally invokes act to run the workflow locally inside a Docker runner. If act is not installed, the script prints a manual verification checklist.
Reference
Full specification: see the canonical spec at ../../../docs/agentskills.io.md.
04 同仓库更多 Skills
validation_handoff
Forces the agent to conclude its response by generating a strict validation checklist and outlining pending actions, focusing heavily on defensive execution and edge-case testing.
project-ployglot-project-scafford
Scaffold a polyglot library project as TypeScript + Python twin packages under `packages/ts/` + `packages/py/`, with a shared `SPEC.md`, cross-language `tests/parity/fixtures.json`, side-by-side `examples/sdk/` + `examples/api/` documentation, per-package `Makefile`, and a root orchestrator that exposes `make ci`. Open-source friendly — no organization-specific paths or scopes baked in. Use when starting a new polyglot library, porting a single-language library to a twin, or auditing/regularizing an existing two-language repo against the canonical layout.
playwright-ui-locator
Auto-heal Playwright tests after a frontend developer changes a div's class, id, or other selector. Catches a failing Playwright test, captures the old vs new DOM tree, finds the element that drifted, derives the most stable replacement locator, and proposes the fix as a GitHub pull request (confident heals) or files a GitHub issue (ambiguous / element-removed cases) — never editing the working tree directly. Trigger when Playwright tests fail with "locator resolved to 0 elements", "Timeout waiting for selector", or after a UI refactor / class rename / Tailwind cleanup broke previously-passing tests.
org-learn-context-window
Distill the current conversation into a reusable command or skill artifact. Use when the user wants to capture a workflow they just executed (create a reusable slash command) or formalize a resolution/mistake/correction into a reusable skill so the same class of issue never recurs.
org-agent-skill-creator
Scaffold and validate scope-aware agent skills (company, enterprise, org, team, project, application). Use when the user asks to create a new agent skill, generate a SKILL.md file, scaffold a skill directory, or validate that existing skill directories meet SKILL_SPEC.md (frontmatter, naming rules, structure, scope/tier).
app-workshop-session-pdf-recap-next
Extract content from a workshop session PDF and produce a structured recap focused on (1) what was learned, (2) next steps, and (3) possible enhancements. Ingests a single PDF file path, parses session material (slides, transcripts, exercises, Q&A), and emits a Markdown report with Learning Recap, Next Steps, and Enhancement Opportunities sections with page-level traceability. Use when the user asks to "summarize a workshop", "recap a training PDF", "pull next steps from a session deck", or provides a workshop/seminar/tutorial PDF and wants a learn-and-act summary.