{"id":"d90a0948-f26f-4340-a6c0-72c2513b671b","shortId":"Yz28sx","kind":"skill","title":"vibe","tagline":"Vibe Code Orchestrator (VCO) is a governed runtime entry that freezes requirements, plans XL-first execution, and enforces verification and phase cleanup.","description":"# Vibe Governed Runtime Entry\n\nThis file is the host-facing SOP for entering canonical `vibe`. Keep it small:\nruntime details belong in `protocols/runtime.md`, execution discipline belongs in\n`protocols/do.md`, and host wrapper recipes belong in installer-generated wrapper\ndocs.\n\n## Trigger Contract\n\nEnter canonical `vibe` before ordinary execution when the user explicitly invokes\n`$vibe`, `/vibe`, or the `vibe` skill, or when the host intentionally chooses\ngoverned requirement/plan/execution closure for a complex task.\n\nDo not route every loosely related task into `vibe`. Lightweight questions,\nsingle-command checks, or tasks better served by another explicitly requested\nskill may proceed outside `vibe` unless the user explicitly invoked this entry.\n\n`vibe-upgrade` is a separate public skill for upgrading the installed\nVibe-Skills project. Do not relaunch an upgrade request as `entry_id = vibe`;\nuse the `vibe-upgrade` skill and its backend instead.\n\nUser instructions remain highest priority. If CLAUDE.md, GEMINI.md, AGENTS.md,\nor the direct user request narrows or forbids a workflow such as TDD, follow the\nuser's instruction while preserving canonical launch and proof rules.\n\n## Canonical Launch SOP\n\nBefore canonical launch, do only the minimum needed to launch:\n\n- Resolve `skill_root`: the directory containing this `SKILL.md`.\n- Resolve `workspace_root`: the task workspace where governed artifacts should\n  be written.\n- Resolve `host_id`: `codex`, `claude-code`, `cursor`, `windsurf`, `openclaw`,\n  or `opencode`.\n- Extract core intent as keyword text. Do not pass the raw prompt, full chat\n  history, or mixed-language filler to the router.\n\nDo not inspect repository files, protocol docs, previous run outputs, or old\nproof artifacts before canonical launch returns. Reading this file, a wrapper,\nor an AGENTS/CLAUDE bootstrap block is not proof of canonical entry.\n\nInternal specialist recommendation router: `scripts/router/resolve-pack-route.ps1`\n\nSpecialist recommender input rules:\n\nThis recommender runs inside canonical `vibe`; it may suggest specialist skills, but it does not decide whether `$vibe` is the public runtime entry.\n\n- Include work type, domain/technology, deliverable, and explicit constraints.\n- Reuse verified frozen requirement/plan facts when continuing a run.\n- If the router returns `confirm_required`, surface the machine-readable route\n  contract and convert the user's natural-language reply into a structured route\n  decision.\n- If the router fails, report `blocked` with the concrete failure reason.\n\nCanonical entry command shape:\n\n```powershell\n$env:PYTHONPATH = \"<skill_root>/apps/vgo-cli/src\"\npy -3 -m vgo_cli.main canonical-entry `\n  --repo-root \"<skill_root>\" `\n  --artifact-root \"<workspace_root>\" `\n  --host-id \"<host_id>\" `\n  --entry-id \"vibe\" `\n  --prompt \"<extracted keyword intent text>\"\n```\n\nBash-like hosts, including Claude Code, should avoid Bash-wrapped PowerShell.\nSet `PYTHONPATH` in the outer shell and call Python directly:\n\n```bash\nREPO_ROOT='<skill_root>'\nWORKSPACE_ROOT=\"${WORKSPACE_ROOT:-$PWD}\"\nPYTHONPATH=\"$REPO_ROOT/apps/vgo-cli/src\" py -3 -m vgo_cli.main canonical-entry \\\n  --repo-root \"$REPO_ROOT\" \\\n  --artifact-root \"$WORKSPACE_ROOT\" \\\n  --host-id \"<host_id>\" \\\n  --entry-id \"vibe\" \\\n  --prompt \"<extracted keyword intent text>\"\n```\n\nAfter canonical-entry returns a `session_root`, validate proof artifacts only\ninside that launched session:\n\n- `host-launch-receipt.json`\n- `runtime-input-packet.json`\n- `governance-capsule.json`\n- `stage-lineage.json`\n\n## Bounded Stop And Re-entry\n\n`vibe` uses progressive governed stops:\n\n1. `requirement_doc`\n2. `xl_plan`\n3. `phase_cleanup`\n\nWhen `bounded_return_control.explicit_user_reentry_required = true`, stop the\ncurrent assistant turn. Do not consume re-entry credentials until a later user\nmessage approves or revises the current boundary.\n\nFor re-entry, inspect `runtime-summary.json ->\nbounded_return_control.host_decision_contract`, infer the user's intent, and\nwrite a structured host decision JSON file. Use the same `run_id`,\n`bounded_reentry_token`, and stable `workspace_root`:\n\n```bash\nREPO_ROOT='<skill_root>'\nWORKSPACE_ROOT=\"${WORKSPACE_ROOT:-$PWD}\"\nDECISION_JSON=\"$WORKSPACE_ROOT/.vibeskills/tmp/host-decision.json\"\nmkdir -p \"$(dirname \"$DECISION_JSON\")\"\n\ncat > \"$DECISION_JSON\" <<'JSON'\n{\n  \"decision_kind\": \"approval_response\",\n  \"decision_action\": \"approve_requirement\",\n  \"approval_decision\": \"approve\"\n}\nJSON\n\nPYTHONPATH=\"$REPO_ROOT/apps/vgo-cli/src\" py -3 -m vgo_cli.main canonical-entry \\\n  --repo-root \"$REPO_ROOT\" \\\n  --artifact-root \"$WORKSPACE_ROOT\" \\\n  --host-id \"<host_id>\" \\\n  --entry-id \"vibe\" \\\n  --prompt \"<stable continuation intent, not just the user's short reply>\" \\\n  --continue-from-run-id \"<source_run_id>\" \\\n  --bounded-reentry-token \"<reentry_token>\" \\\n  --host-decision-json-file \"$DECISION_JSON\"\n```\n\nA structured approval advances to the next progressive stop. A structured\nrevision must include non-empty `revision_delta` and refreezes the same bounded\nstage without asking the user for a separate approval first:\n\n```json\n{\n  \"decision_kind\": \"approval_response\",\n  \"decision_action\": \"revise_requirement\",\n  \"approval_decision\": \"revise\",\n  \"revision_delta\": [\n    \"Freeze one public small/medium face dataset downloaded locally.\",\n    \"Require a polished LaTeX paper and compiled PDF.\"\n  ]\n}\n```\n\nRoute confirmations must stay inside surfaced confirm options. Bounded approvals\nor revisions must stay inside the surfaced bounded-stage action contract.\n\n## Runtime Contract Summary\n\nCanonical `vibe` owns one runtime authority and one visible requirement/plan\nsurface. The fixed state machine is:\n\n1. `skeleton_check`\n2. `deep_interview`\n3. `requirement_doc`\n4. `xl_plan`\n5. `plan_execute`\n6. `phase_cleanup`\n\nThese stages may be light for simple work, but they are not silently skipped.\nThe full runtime contract, stage ownership, lineage rules, internal `M`/`L`/`XL`\ngrades, cleanup rules, and output inventory are defined in\n`protocols/runtime.md`.\n\nPublic wrapper entries remain limited to:\n\n- `vibe`\n- `vibe-upgrade`\n\nCompatibility stage IDs are non-public metadata and must not be materialized as\nhost-visible command or skill wrappers:\n\n- `vibe-what-do-i-want` -> `requirement_doc`\n- `vibe-how-do-we-do` -> `xl_plan`\n- `vibe-do-it` -> `phase_cleanup`\n\n## Skill Execution\n\nThe router may surface selected skill execution candidates, but `vibe` remains\nthe runtime-selected skill and runtime authority.\n\nThe host must inspect surfaced candidates and make a structured skill execution\ndecision when curation is needed. It may approve, defer, or reject only surfaced\ncandidate ids. Unsuitable or noisy candidates should be rejected or deferred\nwith a reason rather than forced into execution.\n\nOnly selected skills become execution units. The host must not invent unsurfaced\nskills, bypass runtime validation, create hidden skill sub-sessions, or open a\nsecond requirement/plan/runtime surface. Selected skill work must preserve the\nskill's own workflow, inputs, outputs, and validation style.\n\nFor XL delegation, root/child hierarchy remains governed: only `root_governed`\nmay freeze canonical requirements/plans or make final completion claims.\n`child_governed` lanes inherit the frozen context, stay inside assigned write\nscopes, validate `delegation-envelope.json`, and emit local receipts only.\n\n## Quality Rules\n\nNever claim success without evidence. Minimum invariants:\n\n- Verify before completion.\n- Do not make silent no-regression claims.\n- Keep requirement and plan artifacts traceable to the launched run.\n- Emit cleanup receipts before claiming phase completion.\n- Expose failures, fallback, degraded status, or blocked state explicitly.\n- Do not add mock success paths, swallowed errors, or template-only pass results.\n- Do not use fallback or boundary behavior to bypass real execution,\n  verification, or root-cause repair.\n\n## Protocol Map\n\nRead these references only after canonical launch or when maintaining the repo:\n\n- `protocols/runtime.md`: governed runtime contract and stage ownership\n- `protocols/think.md`: planning, research, and pre-execution analysis\n- `protocols/do.md`: coding, debugging, and verification\n- `protocols/review.md`: review and quality gates\n- `protocols/team.md`: XL multi-agent orchestration\n- `protocols/retro.md`: retrospective and evidence-backed corrections\n\n## Maintenance\n\n- Runtime family: governed-runtime-first\n- Version: 3.1.1\n- Updated: 2026-05-06\n- Internal specialist recommendation router: `scripts/router/resolve-pack-route.ps1`\n- Primary contract metadata: `core/skill-contracts/v1/vibe.json`","tags":["vibe","skills","foryourhealth111-pixel","agent-framework","agent-skills","agentic-coding","ai-agents","ai-scientist","ai-skills","ai-workflow","automation","claude-code"],"capabilities":["skill","source-foryourhealth111-pixel","skill-vibe-skills","topic-agent-framework","topic-agent-skills","topic-agentic-coding","topic-ai-agents","topic-ai-scientist","topic-ai-skills","topic-ai-workflow","topic-automation","topic-claude-code","topic-codex","topic-context-engineering","topic-developer-tools"],"categories":["Vibe-Skills"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/foryourhealth111-pixel/Vibe-Skills","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add foryourhealth111-pixel/Vibe-Skills","source_repo":"https://github.com/foryourhealth111-pixel/Vibe-Skills","install_from":"skills.sh"}},"qualityScore":"0.700","qualityRationale":"deterministic score 0.70 from registry signals: · indexed on github topic:agent-skills · 2116 github stars · SKILL.md body (8,643 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":"2026-05-18T18:52:54.343Z","embedding":null,"createdAt":"2026-04-18T23:35:01.243Z","updatedAt":"2026-05-18T18:52:54.343Z","lastSeenAt":"2026-05-18T18:52:54.343Z","tsv":"'-05':1167 '-06':1168 '-3':400,455,619 '/apps/vgo-cli/src':398 '/vibe':79 '1':510,774 '2':513,777 '2026':1166 '3':516,780 '3.1.1':1164 '4':783 '5':786 '6':789 'action':608,709,753 'add':1075 'advanc':672 'agent':1147 'agents.md':176 'agents/claude':295 'analysi':1132 'anoth':117 'approv':542,605,609,611,613,671,701,706,712,742,921 'artifact':231,283,410,467,489,631,1051 'artifact-root':409,466,630 'ask':695 'assign':1017 'assist':528 'author':763,901 'avoid':428 'back':1154 'backend':166 'bash':421,430,443,582 'bash-lik':420 'bash-wrap':429 'becom':949 'behavior':1093 'belong':46,51,58 'better':114 'block':297,385,1070 'bootstrap':296 'bound':499,575,659,692,741,751 'boundari':547,1092 'bounded-reentry-token':658 'bounded-stag':750 'bounded_return_control.explicit':520 'bounded_return_control.host':554 'bypass':959,1095 'call':440 'candid':890,907,927,932 'canon':39,68,197,202,206,285,302,317,391,404,459,481,623,758,1001,1111 'canonical-entri':403,458,480,622 'cat':599 'caus':1102 'chat':260 'check':111,776 'child':1008 'choos':89 'claim':1007,1030,1046,1061 'claud':240,425 'claude-cod':239 'claude.md':174 'cleanup':24,518,791,819,880,1058 'closur':92 'code':3,241,426,1134 'codex':238 'command':110,393,855 'compat':838 'compil':731 'complet':1006,1038,1063 'complex':95 'concret':388 'confirm':357,734,739 'constraint':343 'consum':532 'contain':220 'context':1014 'continu':350,644,654 'continue-from-run-id':653 'contract':66,365,556,754,756,809,1121,1175 'convert':367 'core':248 'core/skill-contracts/v1/vibe.json':1177 'correct':1155 'creat':962 'credenti':536 'curat':916 'current':527,546 'cursor':242 'dataset':722 'debug':1135 'decid':328 'decis':379,555,567,590,597,600,603,607,612,664,667,704,708,713,914 'deep':778 'defer':922,937 'defin':825 'degrad':1067 'deleg':991 'delegation-envelope.json':1021 'deliver':340 'delta':687,716 'detail':45 'direct':179,442 'directori':219 'dirnam':596 'disciplin':50 'doc':64,276,512,782,866 'domain/technology':339 'download':723 'emit':1023,1057 'empti':685 'enforc':20 'enter':38,67 'entri':10,28,131,155,303,335,392,405,416,460,475,482,504,535,551,624,639,830 'entry-id':415,474,638 'env':396 'error':1080 'everi':100 'evid':1033,1153 'evidence-back':1152 'execut':18,49,72,788,882,889,913,945,950,1097,1131 'explicit':76,118,128,342,1072 'expos':1064 'extract':247 'face':35,721 'fact':348 'fail':383 'failur':389,1065 'fallback':1066,1090 'famili':1158 'file':30,274,290,569,666 'filler':266 'final':1005 'first':17,702,1162 'fix':770 'follow':190 'forbid':184 'forc':943 'freez':12,717,1000 'frozen':346,1013 'full':259,807 'gate':1142 'gemini.md':175 'generat':62 'govern':8,26,90,230,508,995,998,1009,1119,1160 'governance-capsule.json':497 'governed-runtime-first':1159 'grade':818 'hidden':963 'hierarchi':993 'highest':171 'histori':261 'host':34,55,87,236,413,423,472,566,636,663,853,903,953 'host-decision-json-fil':662 'host-fac':33 'host-id':412,471,635 'host-launch-receipt.json':495 'host-vis':852 'id':156,237,414,417,473,476,574,637,640,657,840,928 'includ':336,424,682 'infer':557 'inherit':1011 'input':311,984 'insid':316,491,737,747,1016 'inspect':272,552,905 'instal':61,143 'installer-gener':60 'instead':167 'instruct':169,194 'intent':88,249,561,645 'intern':304,814,1169 'interview':779 'invari':1035 'invent':956 'inventori':823 'invok':77,129 'json':568,591,598,601,602,614,665,668,703 'keep':41,1047 'keyword':251 'kind':604,705 'l':816 'lane':1010 'languag':265,373 'later':539 'latex':728 'launch':198,203,207,214,286,493,1055,1112 'light':796 'lightweight':106 'like':422 'limit':832 'lineag':812 'local':724,1024 'loos':101 'm':401,456,620,815 'machin':362,772 'machine-read':361 'maintain':1115 'mainten':1156 'make':909,1004,1041 'map':1105 'materi':850 'may':121,320,794,885,920,999 'messag':541 'metadata':845,1176 'minimum':211,1034 'mix':264 'mixed-languag':263 'mkdir':594 'mock':1076 'multi':1146 'multi-ag':1145 'must':681,735,745,847,904,954,977 'narrow':182 'natur':372 'natural-languag':371 'need':212,918 'never':1029 'next':675 'no-regress':1043 'noisi':931 'non':684,843 'non-empti':683 'non-publ':842 'old':281 'one':718,761,765 'open':969 'openclaw':244 'opencod':246 'option':740 'orchestr':4,1148 'ordinari':71 'outer':437 'output':279,822,985 'outsid':123 'own':760 'ownership':811,1124 'p':595 'paper':729 'pass':255,1085 'path':1078 'pdf':732 'phase':23,517,790,879,1062 'plan':14,515,785,787,874,1050,1126 'polish':727 'powershel':395,432 'pre':1130 'pre-execut':1129 'preserv':196,978 'previous':277 'primari':1174 'prioriti':172 'proceed':122 'progress':507,676 'project':147 'prompt':258,419,478,642 'proof':200,282,300,488 'protocol':275,1104 'protocols/do.md':53,1133 'protocols/retro.md':1149 'protocols/review.md':1138 'protocols/runtime.md':48,827,1118 'protocols/team.md':1143 'protocols/think.md':1125 'public':138,333,719,828,844 'pwd':450,589 'py':399,454,618 'python':441 'pythonpath':397,434,451,615 'qualiti':1027,1141 'question':107 'rather':941 'raw':257 're':503,534,550 're-entri':502,533,549 'read':288,1106 'readabl':363 'real':1096 'reason':390,940 'receipt':1025,1059 'recip':57 'recommend':306,310,314,1171 'reentri':522,576,660 'refer':1108 'refreez':689 'regress':1045 'reject':924,935 'relat':102 'relaunch':150 'remain':170,831,893,994 'repair':1103 'repli':374,652 'repo':407,444,452,462,464,583,616,626,628,1117 'repo-root':406,461,625 'report':384 'repositori':273 'request':119,153,181 'requir':13,358,511,523,610,711,725,781,865,1048 'requirement/plan':347,767 'requirement/plan/execution':91 'requirement/plan/runtime':972 'requirements/plans':1002 'research':1127 'resolv':215,223,235 'respons':606,707 'result':1086 'retrospect':1150 'return':287,356,483 'reus':344 'review':1139 'revis':544,680,686,710,714,715,744 'root':217,225,408,411,445,447,449,463,465,468,470,486,581,584,586,588,627,629,632,634,997,1101 'root-caus':1100 'root/.vibeskills/tmp/host-decision.json':593 'root/apps/vgo-cli/src':453,617 'root/child':992 'rout':99,364,378,733 'router':269,307,355,382,884,1172 'rule':201,312,813,820,1028 'run':278,315,352,573,656,1056 'runtim':9,27,44,334,755,762,808,896,900,960,1120,1157,1161 'runtime-input-packet.json':496 'runtime-select':895 'runtime-summary.json':553 'scope':1019 'scripts/router/resolve-pack-route.ps1':308,1173 'second':971 'select':887,897,947,974 'separ':137,700 'serv':115 'session':485,494,967 'set':433 'shape':394 'shell':438 'short':651 'silent':804,1042 'simpl':798 'singl':109 'single-command':108 'skeleton':775 'skill':83,120,139,146,163,216,323,857,881,888,898,912,948,958,964,975,980 'skill-vibe-skills' 'skill.md':222 'skip':805 'small':43 'small/medium':720 'sop':36,204 'source-foryourhealth111-pixel' 'specialist':305,309,322,1170 'stabl':579,643 'stage':693,752,793,810,839,1123 'stage-lineage.json':498 'state':771,1071 'status':1068 'stay':736,746,1015 'stop':500,509,525,677 'structur':377,565,670,679,911 'style':988 'sub':966 'sub-sess':965 'success':1031,1077 'suggest':321 'summari':757 'surfac':359,738,749,768,886,906,926,973 'swallow':1079 'task':96,103,113,227 'tdd':189 'templat':1083 'template-on':1082 'text':252 'token':577,661 'topic-agent-framework' 'topic-agent-skills' 'topic-agentic-coding' 'topic-ai-agents' 'topic-ai-scientist' 'topic-ai-skills' 'topic-ai-workflow' 'topic-automation' 'topic-claude-code' 'topic-codex' 'topic-context-engineering' 'topic-developer-tools' 'traceabl':1052 'trigger':65 'true':524 'turn':529 'type':338 'unit':951 'unless':125 'unsuit':929 'unsurfac':957 'updat':1165 'upgrad':134,141,152,162,837 'use':158,506,570,1089 'user':75,127,168,180,192,369,521,540,559,649,697 'valid':487,961,987,1020 'vco':5 'verif':21,1098,1137 'verifi':345,1036 'version':1163 'vgo_cli.main':402,457,621 'vibe':1,2,25,40,69,78,82,105,124,133,145,157,161,318,330,418,477,505,641,759,834,836,860,868,876,892 'vibe-do-it':875 'vibe-how-do-we-do':867 'vibe-skil':144 'vibe-upgrad':132,160,835 'vibe-what-do-i-w':859 'visibl':766,854 'want':864 'whether':329 'windsurf':243 'without':694,1032 'work':337,799,976 'workflow':186,983 'workspac':224,228,446,448,469,580,585,587,592,633 'wrap':431 'wrapper':56,63,292,829,858 'write':563,1018 'written':234 'xl':16,514,784,817,873,990,1144 'xl-first':15","prices":[{"id":"70245c92-200a-44ba-93b9-ab7b33dcbbcc","listingId":"d90a0948-f26f-4340-a6c0-72c2513b671b","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"foryourhealth111-pixel","category":"Vibe-Skills","install_from":"skills.sh"},"createdAt":"2026-04-18T23:35:01.243Z"}],"sources":[{"listingId":"d90a0948-f26f-4340-a6c0-72c2513b671b","source":"github","sourceId":"foryourhealth111-pixel/Vibe-Skills","sourceUrl":"https://github.com/foryourhealth111-pixel/Vibe-Skills","isPrimary":false,"firstSeenAt":"2026-04-18T23:35:01.243Z","lastSeenAt":"2026-05-18T18:52:54.343Z"}],"details":{"listingId":"d90a0948-f26f-4340-a6c0-72c2513b671b","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"foryourhealth111-pixel","slug":"Vibe-Skills","github":{"repo":"foryourhealth111-pixel/Vibe-Skills","stars":2116,"topics":["agent-framework","agent-skills","agentic-coding","ai-agents","ai-scientist","ai-skills","ai-workflow","automation","claude-code","codex","context-engineering","developer-tools","llm","multi-agent","prompt-engineering","skills","vibe-coding","vibecoding","workflow-automation"],"license":"apache-2.0","html_url":"https://github.com/foryourhealth111-pixel/Vibe-Skills","pushed_at":"2026-05-08T04:21:24Z","description":"Vibe-Skills is an all-in-one AI skills package. It seamlessly integrates expert-level capabilities and context management into a general-purpose skills package， enabling any AI agent to instantly upgrade its functionality—eliminating the friction of fragmented tools and complex harnesses.","skill_md_sha":"36f40a19dd075a20056ebe221bcbec2f2bf1dbb2","skill_md_path":"SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/foryourhealth111-pixel/Vibe-Skills"},"layout":"root","source":"github","category":"Vibe-Skills","frontmatter":{"name":"vibe","description":"Vibe Code Orchestrator (VCO) is a governed runtime entry that freezes requirements, plans XL-first execution, and enforces verification and phase cleanup."},"skills_sh_url":"https://skills.sh/foryourhealth111-pixel/Vibe-Skills"},"updatedAt":"2026-05-18T18:52:54.343Z"}}