{
  "markdown": "# Agent Skills\n\n[Agent skills](https://agentskills.io) that work across coding agents.\n\n## Install\n\n| Agent | Command |\n| --- | --- |\n| Pi | `pi install https://github.com/sheurich/agent-skills` |\n| Claude Code | `/plugin marketplace add sheurich/agent-skills` |\n| Gemini CLI | `gemini extensions install sheurich/agent-skills` |\n| npx skills | `npx skills add sheurich/agent-skills --all -g` |\n\nInstall a single skill with npx:\n\n```bash\nnpx skills add sheurich/agent-skills -s <skill-name> -g\n```\n\n## Skills\n\n| Skill | Description |\n| --- | --- |\n| [readiness-review](skills/readiness-review/SKILL.md) | Structured readiness assessment before proceeding to the next environment or phase. |\n| [semgrep-scan](skills/semgrep-scan/SKILL.md) | Scan changed files with Semgrep for security and correctness issues before committing. |\n| [skill-authoring](skills/skill-authoring/SKILL.md) | Write and refine agent skills. Use when generating skills from docs, extracting from sessions, designing tool interfaces, managing lifecycle, or writing for dual-harness patterns. |\n| [swival](skills/swival/SKILL.md) | Delegate tasks to Swival for self-reviewed code changes, secret-safe operations, cached analysis, and A2A agent serving. |\n\n## Plugins\n\n| Plugin | Description |\n| --- | --- |\n| [auto-summarize](plugins/auto-summarize/) | Rolling session summary and automatic session naming for Pi. |\n\n## Pi Packages\n\n| Package | Description |\n| --- | --- |\n| [swival-subagent](packages/swival-subagent/) | Pi extension that delegates tasks to a swival subprocess (reviewer loop, AgentFS sandbox, secret encryption). |\n\n## Contributing\n\nSee [CONTRIBUTING.md](CONTRIBUTING.md) for how to add skills, write tests,\nand run validation.\n\n## Repo Structure\n\n```text\nskills/                        # Simple skills (SKILL.md only)\nplugins/                       # Claude Code plugins (own manifests, hooks, LSP, MCP)\npackages/                      # Pi packages (pi.extensions in package.json)\ntests/scenarios/               # One scenario per skill or package\nscripts/validate-extra.sh      # Repo-specific validation checks\nvendor/agent-validate/         # Vendored validator\nvalidate.sh                    # Thin wrapper → vendor/agent-validate\n.claude-plugin/marketplace.json\ngemini-extension.json\n```\n\n## Validation\n\n```bash\n./validate.sh              # run all checks\n./validate.sh --skip json  # skip specific checks\n```\n\nCI runs [agent-validate](https://github.com/sheurich/agent-validate) on\nevery push and pull request.\n\n## License\n\n[MIT](LICENSE)\n",
  "bytes": 2536,
  "sha": "655ff7b76ac6a0fdd5779ebdb603a33551173141976d835b1e89c8f44755cac9",
  "repo_slug": "sheurich/agent-skills",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/plg_sheurich_agent_skills_7db10b16/readme"
}