{
  "markdown": "# srinitude skills\n\nPortable Agent Skills from one canonical source tree.\n\nThis repository is an [Agent Plugins 1.0.0](https://agent-plugins.org/specification) package of portable skills that follow the [Agent Skills specification](https://agentskills.io/specification), plus a read-only local MCP server.\n\n[![Install with skills.sh](https://img.shields.io/badge/skills.sh-install-111111)](https://www.skills.sh/srinitude)\n\n## What ships\n\n- Canonical skill packages under [`skills/`](skills/).\n- Root [`plugin.json`](plugin.json) and [`mcp.json`](mcp.json) for Agent Plugins v1 clients.\n- Trigger, behavior, failure, recovery, and speed evaluations.\n- A bundled, read-only local MCP server.\n- Existing native plugin metadata and adapters for client-specific routes.\n- Strict JSON Schemas for manifests, reports, benchmarks, and checkpoints.\n\nNo client integration contains a second skill body.\n\nGitHub release versions and skill metadata versions are independent. Skills don't depend on each other's versions.\n\n## Agent Plugins v1\n\nCompatible clients discover the root `plugin.json`, every immediate child in `skills/`, and the root `mcp.json`. Installation, permissions, and trust remain client-controlled.\n\nSee [Agent Plugins v1](docs/agent-plugins.md) for the fixed package layout, current client registry, validation evidence, schema provenance, and security boundary. `PD-101`: that page owns portable-package operating detail, loads for Agent Plugins work, and links back here.\n\n## Install\n\n```sh\nnpx skills add srinitude/skills\n```\n\nUse `npx skills add srinitude/skills --list` to inspect available skills first. The CLI reports anonymous install telemetry to skills.sh unless `DISABLE_TELEMETRY=1` is set.\n\nClone the repository and run `mise run bootstrap` plus `mise run build-mcp` when developing the package. Node 24 or newer is required. [Mise](https://mise.jdx.dev/) pins the full local toolchain.\n\n## Client support\n\n| Client                               | Route                              | Local MCP   |\n| ------------------------------------ | ---------------------------------- | ----------- |\n| MCP-capable Agent Plugins v1 clients | Root portable package              | Yes         |\n| Claude Code                          | Root plugin and marketplace        | Yes         |\n| Codex                                | Root plugin and marketplace        | Yes         |\n| ChatGPT                              | Codex plugin format                | Yes         |\n| Gemini CLI                           | Root extension                     | Yes         |\n| Cursor                               | Agent Plugins bundle               | Yes         |\n| OpenClaw                             | Codex-compatible bundle            | Yes         |\n| Hermes Agent                         | Root Python plugin                 | Skills only |\n| opencode                             | Project config plus skills install | Yes         |\n| Continue                             | Skills CLI adapter                 | Optional    |\n| Aider                                | Read-only config                   | No          |\n\n### Claude Code\n\n```text\n/plugin marketplace add srinitude/skills\n/plugin install srinitude-skills@srinitude-skills\n```\n\nThe plugin loads [`./skills`](skills/) and [`.mcp.json`](.mcp.json) from the same checkout.\n\n### Codex and ChatGPT\n\nStart Codex with plugin support, open `/plugins`, add `srinitude/skills` as a marketplace, and install `srinitude-skills`.\n\n```sh\ncodex --enable plugins\n```\n\nThe same package metadata is the ChatGPT plugin route documented by the Codex plugin format.\n\n### Gemini CLI\n\n```sh\ngemini extensions install https://github.com/srinitude/skills\n```\n\n[`gemini-extension.json`](gemini-extension.json) starts the bundled MCP server from the installed extension path.\n\n### Cursor\n\nLoad the checkout as a local root Agent Plugin. Cursor reads [`plugin.json`](plugin.json), discovers the canonical [`skills/`](skills/) tree, and starts the read-only MCP route from [`mcp.json`](mcp.json). The [`.cursor-plugin/plugin.json`](.cursor-plugin/plugin.json) manifest keeps Cursor-format compatibility without owning a second skill or MCP definition. This repository does not claim a Cursor Marketplace listing.\n\n### OpenClaw\n\nUse the tag-pinned compatible bundle route and cold-discovery check in the [OpenClaw adapter note](adapters/openclaw/README.md).\n\n### Hermes Agent\n\nUse the Skills Hub tap or optional Python plugin route in the [Hermes Agent adapter note](adapters/hermes-agent/README.md). Both load the existing SKILL.md path.\n\n### opencode\n\nUse the repository installation above. [`opencode.json`](opencode.json) starts the local MCP server when running opencode from this checkout.\n\n### Continue and Aider\n\nUse the [Continue adapter note](adapters/continue/README.md) or [Aider adapter note](adapters/aider/README.md). Both point to the canonical SKILL.md rather than copying it.\n\n## Local MCP server\n\nBuild it with:\n\n```sh\nmise run build-mcp\n```\n\nThe stdio server exposes six read-only tools:\n\n- `list_skills`\n- `search_skills`\n- `get_skill`\n- `get_reference`\n- `get_eval_manifest`\n- `validate_skill`\n\nPaths are confined to the repository skill tree. Absolute paths, traversal, hidden files, nested references, and symlink escapes fail closed. The server has no write tool, telemetry, credentials, or network call.\n\n## Validate and evaluate\n\n```sh\nSKILL_NAME=your-skill-name\nmise run validate -- --all --report .artifacts/skill-validation.json\nmise run eval -- --skill \"$SKILL_NAME\" --transport fixture --report \".artifacts/evals/$SKILL_NAME-fixture\"\nmise run benchmark -- --skill \"$SKILL_NAME\" --transport fixture --samples 1000 --report \".artifacts/benchmarks/$SKILL_NAME-fixture.json\"\n```\n\nFixture results prove runner behavior only. They are not evidence about a language model. Paid OpenRouter evaluation is a separate post-release step with a frozen model inventory, checkpoint files, a cost estimate, and explicit spend approval.\n\nUse the fixed-route, capped procedure in [OpenRouter sweeps](docs/openrouter-sweeps.md).\n\nRun the complete local gate with:\n\n```sh\nmise run ci\n```\n\n## Contribute and report problems\n\nRead [CONTRIBUTING.md](CONTRIBUTING.md) before changing a skill or integration. The [skills.sh publishing notes](docs/skills-sh.md) record discovery, telemetry, API, and listing-correction rules. Security reports follow [SECURITY.md](SECURITY.md). Other help is covered by [SUPPORT.md](SUPPORT.md).\n\n## License\n\n[MIT](LICENSE)\n",
  "bytes": 6504,
  "sha": "9e0f0b1ab85a8f706233dd0b3dd2684b1bdfb737a03f59d51ace1b35a1e7d179",
  "repo_slug": "srinitude/skills",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/plg_srinitude_skills_83d02a08/readme"
}