{"id":50395077,"url":"https://github.com/hsnice16/agent-friendly-skill","last_synced_at":"2026-05-30T20:30:38.325Z","repository":{"id":355869064,"uuid":"1225889891","full_name":"hsnice16/agent-friendly-skill","owner":"hsnice16","description":"Portable agent skill that scores the current repo's agent-friendliness on disk and recommends a model class to use for it.","archived":false,"fork":false,"pushed_at":"2026-05-05T15:25:21.000Z","size":103,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-05-05T17:27:39.387Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"TypeScript","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/hsnice16.png","metadata":{"files":{"readme":"README.md","changelog":"CHANGELOG.md","contributing":"CONTRIBUTING.md","funding":".github/FUNDING.yml","license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":".github/CODEOWNERS","security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":"AGENTS.md","dco":null,"cla":null},"funding":{"github":"hsnice16"}},"created_at":"2026-04-30T18:49:00.000Z","updated_at":"2026-05-05T15:25:20.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/hsnice16/agent-friendly-skill","commit_stats":null,"previous_names":["hsnice16/agent-friendly-skill"],"tags_count":4,"template":false,"template_full_name":null,"purl":"pkg:github/hsnice16/agent-friendly-skill","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hsnice16%2Fagent-friendly-skill","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hsnice16%2Fagent-friendly-skill/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hsnice16%2Fagent-friendly-skill/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hsnice16%2Fagent-friendly-skill/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/hsnice16","download_url":"https://codeload.github.com/hsnice16/agent-friendly-skill/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hsnice16%2Fagent-friendly-skill/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":33709269,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-05-30T02:00:06.278Z","response_time":92,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2026-05-30T20:30:36.683Z","updated_at":"2026-05-30T20:30:38.307Z","avatar_url":"https://github.com/hsnice16.png","language":"TypeScript","funding_links":["https://github.com/sponsors/hsnice16"],"categories":[],"sub_categories":[],"readme":"# agent-friendly-skill\n\nPortable agent skill that scores the current repo's agent-friendliness on disk and recommends a model class to use for it. Profiles eight agents — **Claude Code, Cursor, Devin, GPT-5 Codex, Gemini CLI, Aider, OpenHands, and Pi** — the same set the [Agent Friendly Code](https://github.com/hsnice16/agent-friendly-code) dashboard scores against. Installs into any [`vercel-labs/skills`](https://github.com/vercel-labs/skills)-compatible agent; agents without a profile (Cline, Copilot, Continue, Roo Code, …) still install and run fine — the recommendation is score-driven and provider-neutral, so the overall score and the score → model-class mapping apply regardless of which agent invoked the skill.\n\n\u003e **Version**: `v0.1.3` — plans live in [`tasks/`](./tasks/). Pin `#v0` to track the latest 0.x release; pin a precise tag (`#v0.1.3`) to opt out of automatic minor/patch updates — see [Pinning a version](#pinning-a-version) below. Tracking task in the parent project: [`tasks/0.5.0/03-agent-skill.md`](https://github.com/hsnice16/agent-friendly-code/blob/main/tasks/0.5.0/03-agent-skill.md).\n\n## What it does\n\nWhen the user invokes `/agent-friendly` (or however their agent triggers skills), the skill:\n\n1. Tells the user upfront that it scores the current working directory — so they're at the project root, or pass an explicit path. The agent surfaces this; the CLI also emits a `warnings[]` entry when the cwd has no project markers (`package.json` / `README.md` / `AGENTS.md` / `.git`) so a wrong path can't silently produce a low score.\n2. Runs the bundled scorer locally — 16 static signals × 8 agent profiles.\n3. Picks the highest-scoring agent as the **best-fit** for this repo (score-driven, regardless of which agent invoked the skill).\n4. Maps the overall score to a band (high / mid / low) and recommends a model **class** (frontier / standard / small) for the user to switch to manually — the skill never switches the model programmatically.\n5. Prints the overall score, best-fit agent, runner-up if close, top improvements.\n\nThe scorer is **vendored from the upstream [`agent-friendly-code`](https://github.com/hsnice16/agent-friendly-code) repo** and bundled into `skills/agent-friendly/dist/index.js` via `@vercel/ncc`. The skill never makes an HTTP request — if the upstream dashboard goes offline tomorrow, this skill keeps producing recommendations unchanged.\n\n## Install\n\n```bash\nnpx skills add hsnice16/agent-friendly-skill\n```\n\nThe [`vercel-labs/skills`](https://github.com/vercel-labs/skills) CLI autodetects which agents are configured locally and writes `SKILL.md` plus `dist/index.js` to each one's skill directory. Pass `--agent \u003cname\u003e` to scope, or `--agent '*'` for every detected agent. Default install is a symlink to a canonical copy, so `npx skills update` propagates to every agent at once.\n\nAfter install, ask your agent something like _\"is this repo agent-friendly?\"_ or invoke `/agent-friendly` directly.\n\n## Pinning a version\n\nThe `vercel-labs/skills` CLI uses `#\u003cref\u003e` (a URL fragment) to pick a git ref — **not** `@\u003cref\u003e`. The `@` syntax is reserved for filtering by skill name in multi-skill repos.\n\n```bash\nnpx skills add hsnice16/agent-friendly-skill          # tracks main — auto-updates via `npx skills update`\nnpx skills add hsnice16/agent-friendly-skill#v0       # floating major tag — auto-updates within 0.x.y\nnpx skills add hsnice16/agent-friendly-skill#v0.1.3   # precise tag — never updates\n```\n\nBranches and tags are both accepted. Commit SHAs are not (the CLI clones with `--branch \u003cref\u003e` under the hood, which doesn't accept raw SHAs).\n\n## Score → model mapping\n\n| Band | Score   | Recommendation                                                                                         |\n| ---- | ------- | ------------------------------------------------------------------------------------------------------ |\n| High | ≥ 80    | Frontier — Opus / GPT-5 / Gemini 2.5 Pro. Repo is well-prepped, the model can leverage it.             |\n| Mid  | 60 – 79 | Standard — Sonnet / GPT-5 Codex / Gemini 2.5 Flash. Solid baseline; frontier optional.                 |\n| Low  | \u003c 60    | Small / fast — Haiku / GPT-4o-mini / Gemini 2.5 Flash-Lite. Repo lacks scaffolding for a frontier run. |\n\nProvider-neutral. The skill recommends a model **class** — the user picks the actual ID for their agent.\n\n## Optional — auto-score every session\n\nFor agents that support session-start hooks, drop the matching snippet from \u003chttps://www.agentfriendlycode.com/skill\u003e into your settings file. The hook runs `node \u003cskill-dir\u003e/dist/index.js . --summary` and prints a one-line score at the top of every session.\n\n## Usage outside the skill\n\nThe bundled CLI also runs standalone:\n\n```bash\nnode skills/agent-friendly/dist/index.js .            # JSON output (default)\nnode skills/agent-friendly/dist/index.js . --summary  # one-line human summary\n```\n\nUseful in CI checks, status-line scripts, or anywhere you want a portable agent-friendliness score without installing the skill.\n\n## Self-contained\n\nThe scorer + weights live inside `skills/agent-friendly/dist/index.js`. No network call to the upstream dashboard at runtime. The vendored scorer stays in sync with upstream by hand — see [AGENTS.md](./AGENTS.md) \"Vendored scorer\" for the propagation rule.\n\n## Family\n\nThree repos, one scorer:\n\n- **[`agent-friendly-code`](https://github.com/hsnice16/agent-friendly-code)** — the public dashboard at \u003chttps://www.agentfriendlycode.com\u003e, source of truth for the scorer and the per-model weights.\n- **[`agent-friendly-action`](https://github.com/hsnice16/agent-friendly-action)** — GitHub Action that scores PR head + base inside maintainer CI and posts a single delta comment on every PR. Same vendored scorer, different surface.\n- **`agent-friendly-skill`** (this repo) — local in-editor counterpart for any [`vercel-labs/skills`](https://github.com/vercel-labs/skills)-compatible agent.\n\nWhen the upstream scorer changes, the action and skill mirror by hand — see each repo's `AGENTS.md` \"Vendored scorer\" section for the propagation rule.\n\n## Changelog\n\nSee [`CHANGELOG.md`](./CHANGELOG.md) for what shipped per release.\n\n## Sponsor\n\nIf this skill is useful to you, consider sponsoring its development: [github.com/sponsors/hsnice16](https://github.com/sponsors/hsnice16).\n\n## License\n\n[MIT](./LICENSE)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fhsnice16%2Fagent-friendly-skill","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fhsnice16%2Fagent-friendly-skill","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fhsnice16%2Fagent-friendly-skill/lists"}