{"id":"7f3eea1c-9fd8-4ecd-b7a5-ff4e235875de","shortId":"q4HXhV","kind":"skill","title":"agent-onboarding","tagline":"Make your repository AI-ready with single skill run","description":"## Purpose\r\nInitialize a repository so that AI coding agents (Claude Code, Codex, Cursor, etc.)\r\ncan operate safely, deterministically, and with minimal human supervision.\r\n\r\nThis skill generates:\r\n- AGENTS.md (agent operational contract)\r\n- docs/ directory with structured technical documentation\r\n\r\n## Inputs\r\n- Repository to initialize.\r\n- Optional: short human note describing the project and instructions (1–3 sentences)\r\n\r\n## Outputs\r\n- AGENTS.md at repository root\r\n- docs/ directory with standardized markdown files\r\n\r\n## Core Principles\r\n- Do NOT guess. If information is missing, explicitly mark TODOs.\r\n- Prefer correctness and safety over completeness.\r\n- Optimize for agent execution, not human prose.\r\n- Minimize exploration cost and failure modes.\r\n- If the given repository follows a monorepo pattern, create a subdirectory under docs/ for each project so the documentation doesn’t get mixed together.\r\n  - Example) `docs/frontend/00_overview.md` `docs/frontend/adr` `docs/frontend/plan`\r\n\r\n---\r\n\r\n## Execution Steps\r\n\r\n### Step 1: Repository Inspection\r\n\r\n- Detect:\r\n  - Primary language(s)\r\n  - Framework(s)\r\n  - Package manager\r\n  - Test framework\r\n  - Deployment indicators (Dockerfile, CI configs, Vercel, etc.)\r\n- Identify:\r\n  - Entry points\r\n  - Core business logic directories\r\n  - Infrastructure-related files (DB, queues, storage)\r\n\r\nIf detection is ambiguous:\r\n- Record ambiguity explicitly in docs/00_overview.md\r\n- Do NOT assume defaults\r\n\r\n---\r\n\r\n### Step 2: Confirm Ambiguities & Decisions With the Human\r\n\r\nRequest the user to provide the following information:\r\n\r\n- Whether Git-related commands are allowed: use freely vs use only with explicit approval\r\n- Language preferences:\r\n  - Language for comments, documents and commit messages\r\n  - Communication language: same as the user vs a specific language\r\n- Any other items discovered in Step 1 that are ambiguous or seem to require questions\r\n\r\nIf the human does not answer:\r\n- Do NOT guess.\r\n- Pause and ask again as needed; if you must proceed, mark missing items as “Unknown / TODO” in docs (without embedding the unanswered questions), and keep AGENTS.md conservative (minimal privileges, no destructive ops).\r\n\r\n---\r\n\r\n### Step 3: Generate AGENTS.md\r\nFirst, copy [AGENTS.md.example](`assets/AGENTS.md.example`), then edit it based on the information gathered so far to create `AGENTS.md`.\r\n\r\nAGENTS.md MUST include:\r\n# Repository Guidelines\r\n## Project Structure & Module Organization\r\n\r\n## Documentation\r\n\r\n## Coding Style & Naming Conventions\r\n\r\n## Commit Message & PR Guidelines\r\n### Commit Message Format\r\n### Atomic Commits\r\n\r\n## Agent Operation & Safety Rules\r\n\r\n## Language Policy\r\n\r\n## User Interaction\r\n\r\nHard rules:\r\n- Keep AGENTS.md concise (target: 50–120 lines)\r\n- Link to docs/ instead of duplicating explanations\r\n\r\n---\r\n\r\n### Step 4: Generate docs/ Structure\r\n\r\n**If the `docs/` directory already exists, generate all new files inside `docs/agents/` instead of `docs/`. Otherwise, use `docs/` as the target directory.**\r\n\r\nCreate the following files if applicable:\r\n\r\n- docs/00_overview.md\r\n- docs/10_architecture.md\r\n- docs/20_dev_env.md\r\n- docs/30_commands.md\r\n- docs/40_testing.md\r\n- docs/50_deploy.md\r\n- docs/60_observability.md\r\n- docs/70_security.md\r\n- docs/80_style_guide.md\r\n- docs/90_troubleshooting.md\r\n\r\nRules:\r\n- Omit files that are clearly irrelevant\r\n- Never fabricate cloud providers, CI tools, or databases\r\n- Prefer explicit “Unknown / TODO” over assumptions\r\n\r\n---\r\n\r\n### Step 5: ADR Detection (Optional but Recommended)\r\nIf architectural decisions are inferred:\r\n- Create docs/adr/\r\n- Record assumptions as provisional ADRs\r\n- Mark them as “UNVERIFIED”\r\n\r\n---\r\n\r\n### Step 6: Validation Checklist\r\nBefore finalizing:\r\n- All commands copy-paste runnable OR marked TODO\r\n- No secrets included\r\n- No speculative tech choices\r\n- AGENTS.md links resolve correctly\r\n\r\n---\r\n\r\n## Failure Handling\r\nIf the repository is:\r\n- Empty\r\n- Too small to infer structure\r\n- Highly ambiguous\r\n\r\nThen:\r\n- Generate minimal AGENTS.md\r\n- Generate docs/00_overview.md that clearly separates detected facts vs “Unknown / TODO” items (but do not include a question list; ask any needed questions during execution).\r\n\r\nIf, at any point, you encounter ambiguities or actions that appear potentially risky:\r\n\r\nThen:\r\n- always pause and request clarification from the user before proceeding.\r\n\r\n---\r\n\r\n## Non-Goals\r\n- Do not optimize for marketing or onboarding humans\r\n- Do not refactor code\r\n- Do not introduce new dependencies\r\n- Do not modify runtime behavior\r\n\r\n---\r\n\r\n## Success Criteria\r\nA coding agent can:\r\n- Identify where to make changes\r\n- Execute tests confidently\r\n- Avoid destructive or unsafe operations\r\n- Complete scoped tasks without human correction","tags":["agent","onboarding","skyil7","agent-skills","agentic-coding","agents","claude-code","codex","skill"],"capabilities":["skill","source-skyil7","skill-agent-onboarding","topic-agent-skills","topic-agentic-coding","topic-agents","topic-claude-code","topic-codex","topic-skill"],"categories":["agent-onboarding"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/skyil7/agent-onboarding","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add skyil7/agent-onboarding","source_repo":"https://github.com/skyil7/agent-onboarding","install_from":"skills.sh"}},"qualityScore":"0.454","qualityRationale":"deterministic score 0.45 from registry signals: · indexed on github topic:agent-skills · 9 github stars · SKILL.md body (4,763 chars)","verified":false,"liveness":"unknown","lastLivenessCheck":null,"agentReviews":{"count":0,"score_avg":null,"cost_usd_avg":null,"success_rate":null,"latency_p50_ms":null,"narrative_summary":null,"summary_updated_at":null},"enrichmentModel":"deterministic:skill-github:v1","enrichmentVersion":1,"enrichedAt":"2026-05-18T19:08:45.002Z","embedding":null,"createdAt":"2026-05-18T13:14:23.040Z","updatedAt":"2026-05-18T19:08:45.002Z","lastSeenAt":"2026-05-18T19:08:45.002Z","tsv":"'1':63,139,242 '120':351 '2':187 '3':64,293 '4':361 '5':425 '50':350 '6':448 'action':523 'adr':426,442 'agent':2,22,41,97,336,568 'agent-onboard':1 'agents.md':40,67,285,295,312,313,347,469,490 'agents.md.example':298 'ai':8,20 'ai-readi':7 'allow':208 'alreadi':369 'alway':529 'ambigu':176,178,189,245,486,521 'answer':256 'appear':525 'applic':392 'approv':216 'architectur':432 'ask':262,509 'assets/agents.md.example':299 'assum':184 'assumpt':423,439 'atom':334 'avoid':578 'base':303 'behavior':563 'busi':163 'chang':574 'checklist':450 'choic':468 'ci':155,414 'clarif':533 'claud':23 'clear':408,494 'cloud':412 'code':21,24,323,553,567 'codex':25 'command':206,454 'comment':221 'commit':224,327,331,335 'communic':226 'complet':94,583 'concis':348 'confid':577 'config':156 'confirm':188 'conserv':286 'contract':43 'convent':326 'copi':297,456 'copy-past':455 'core':77,162 'correct':90,472,588 'cost':104 'creat':116,311,387,436 'criteria':565 'cursor':26 'databas':417 'db':170 'decis':190,433 'default':185 'depend':558 'deploy':152 'describ':58 'destruct':290,579 'detect':142,174,427,496 'determinist':31 'directori':45,72,165,368,386 'discov':239 'doc':44,71,120,277,355,363,367,379,382 'dockerfil':154 'docs/00_overview.md':181,393,492 'docs/10_architecture.md':394 'docs/20_dev_env.md':395 'docs/30_commands.md':396 'docs/40_testing.md':397 'docs/50_deploy.md':398 'docs/60_observability.md':399 'docs/70_security.md':400 'docs/80_style_guide.md':401 'docs/90_troubleshooting.md':402 'docs/adr':437 'docs/agents':376 'docs/frontend/00_overview.md':133 'docs/frontend/adr':134 'docs/frontend/plan':135 'document':49,126,222,322 'doesn':127 'duplic':358 'edit':301 'embed':279 'empti':479 'encount':520 'entri':160 'etc':27,158 'exampl':132 'execut':98,136,514,575 'exist':370 'explan':359 'explicit':86,179,215,419 'explor':103 'fabric':411 'fact':497 'failur':106,473 'far':309 'file':76,169,374,390,405 'final':452 'first':296 'follow':112,200,389 'format':333 'framework':146,151 'freeli':210 'gather':307 'generat':39,294,362,371,488,491 'get':129 'git':204 'git-rel':203 'given':110 'goal':541 'guess':81,259 'guidelin':317,330 'handl':474 'hard':344 'high':485 'human':35,56,100,193,253,549,587 'identifi':159,570 'includ':315,464,505 'indic':153 'infer':435,483 'inform':83,201,306 'infrastructur':167 'infrastructure-rel':166 'initi':15,53 'input':50 'insid':375 'inspect':141 'instead':356,377 'instruct':62 'interact':343 'introduc':556 'irrelev':409 'item':238,272,501 'keep':284,346 'languag':144,217,219,227,235,340 'line':352 'link':353,470 'list':508 'logic':164 'make':4,573 'manag':149 'mark':87,270,443,460 'markdown':75 'market':546 'messag':225,328,332 'minim':34,102,287,489 'miss':85,271 'mix':130 'mode':107 'modifi':561 'modul':320 'monorepo':114 'must':268,314 'name':325 'need':265,511 'never':410 'new':373,557 'non':540 'non-goal':539 'note':57 'omit':404 'onboard':3,548 'op':291 'oper':29,42,337,582 'optim':95,544 'option':54,428 'organ':321 'otherwis':380 'output':66 'packag':148 'past':457 'pattern':115 'paus':260,530 'point':161,518 'polici':341 'potenti':526 'pr':329 'prefer':89,218,418 'primari':143 'principl':78 'privileg':288 'proceed':269,538 'project':60,123,318 'prose':101 'provid':198,413 'provision':441 'purpos':14 'question':250,282,507,512 'queue':171 'readi':9 'recommend':430 'record':177,438 'refactor':552 'relat':168,205 'repositori':6,17,51,69,111,140,316,477 'request':194,532 'requir':249 'resolv':471 'riski':527 'root':70 'rule':339,345,403 'run':13 'runnabl':458 'runtim':562 'safe':30 'safeti':92,338 'scope':584 'secret':463 'seem':247 'sentenc':65 'separ':495 'short':55 'singl':11 'skill':12,38 'skill-agent-onboarding' 'small':481 'source-skyil7' 'specif':234 'specul':466 'standard':74 'step':137,138,186,241,292,360,424,447 'storag':172 'structur':47,319,364,484 'style':324 'subdirectori':118 'success':564 'supervis':36 'target':349,385 'task':585 'tech':467 'technic':48 'test':150,576 'todo':88,275,421,461,500 'togeth':131 'tool':415 'topic-agent-skills' 'topic-agentic-coding' 'topic-agents' 'topic-claude-code' 'topic-codex' 'topic-skill' 'unansw':281 'unknown':274,420,499 'unsaf':581 'unverifi':446 'use':209,212,381 'user':196,231,342,536 'valid':449 'vercel':157 'vs':211,232,498 'whether':202 'without':278,586","prices":[{"id":"8550cd17-87c1-43b7-a946-e7c0ae6e1b71","listingId":"7f3eea1c-9fd8-4ecd-b7a5-ff4e235875de","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"skyil7","category":"agent-onboarding","install_from":"skills.sh"},"createdAt":"2026-05-18T13:14:23.040Z"}],"sources":[{"listingId":"7f3eea1c-9fd8-4ecd-b7a5-ff4e235875de","source":"github","sourceId":"skyil7/agent-onboarding","sourceUrl":"https://github.com/skyil7/agent-onboarding","isPrimary":false,"firstSeenAt":"2026-05-18T13:14:23.040Z","lastSeenAt":"2026-05-18T19:08:45.002Z"}],"details":{"listingId":"7f3eea1c-9fd8-4ecd-b7a5-ff4e235875de","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"skyil7","slug":"agent-onboarding","github":{"repo":"skyil7/agent-onboarding","stars":9,"topics":["agent-skills","agentic-coding","agents","claude-code","codex","skill"],"license":"apache-2.0","html_url":"https://github.com/skyil7/agent-onboarding","pushed_at":"2026-02-02T08:23:48Z","description":"Make your repository AI-ready with single skill run","skill_md_sha":"c5f946ebff6cad90aec9d27ce0a3817654d1a7cb","skill_md_path":"SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/skyil7/agent-onboarding"},"layout":"root","source":"github","category":"agent-onboarding","frontmatter":{},"skills_sh_url":"https://skills.sh/skyil7/agent-onboarding"},"updatedAt":"2026-05-18T19:08:45.002Z"}}