{"id":"05b7b51a-5794-4605-8cd8-fc31c98b2f7c","shortId":"PRZmRK","kind":"skill","title":"closed-loop-delivery","tagline":"Use when a coding task must be completed against explicit acceptance criteria with minimal user re-intervention across implementation, review feedback, deployment, and runtime verification.","description":"# Closed-Loop Delivery\n\n## Overview\n\nTreat each task as incomplete until acceptance criteria are verified in evidence, not until code is merely changed.\n\nCore rule: **deliver against DoD (Definition of Done), not against code diff size.**\n\n## When to Use\nUse this skill when:\n- user gives a coding/fix task and expects end-to-end completion\n- task spans code + tests + PR comments + dev deploy + runtime checks\n- repeated manual prompts like \"now test\", \"now deploy\", \"now re-check PR\" should be avoided\n\nDo not use this skill for:\n- pure Q&A/explanations\n- prod deploy requests without explicit human approval\n- tasks blocked by missing secrets/account access that cannot be inferred\n\n## Required Inputs\n\nBefore execution, define these once:\n- task goal\n- acceptance criteria (DoD)\n- target environment (`dev` by default)\n- max iteration rounds (default `2`)\n\nIf acceptance criteria are missing, request them once. If user does not provide, propose a concrete default and proceed.\n\n## Issue Gate Dependency\n\nBefore execution, prefer using `create-issue-gate`.\n\n- If issue status is `ready` and execution gate is `allowed`, continue.\n- If issue status is `draft`, do not execute implementation/deploy/review loops.\n- Require user-provided, testable acceptance criteria before starting execution.\n\n## Default Workflow\n\n1. **Define DoD**\n   - Convert request into testable criteria.\n   - Example: checkout task DoD = \"checkout endpoint returns a valid, openable third-party payment URL in dev\".\n\n2. **Implement minimal change**\n   - Keep scope tight to task goal.\n\n3. **Verify locally**\n   - Run focused tests first, then broader checks if needed.\n\n4. **Review loop**\n   - Fetch PR comments/reviews.\n   - Classify valid vs non-actionable.\n   - Fix valid items, re-run verification.\n\n5. **Dev deploy + runtime verification**\n   - Deploy to `dev` when runtime behavior matters.\n   - Verify via real API/Lambda/log evidence against DoD.\n\n6. **Completion decision**\n   - Only report \"done\" when all DoD checks pass.\n   - Otherwise continue loop until pass or stop condition.\n\n## PR Comment Polling Policy\n\nAvoid noisy short polling by default. Use batched windows:\n\n- **Round 1:** wait `3m`, collect delta comments/reviews\n- **Round 2:** wait `6m`, collect delta again\n- **Final round:** wait `10m`, collect all remaining visible comments/reviews\n\nAt each round:\n- process all new comments in one batch\n- avoid immediate re-poll after each single comment\n- after the `10m` round, stop waiting and proceed with all comments visible at that point\n\nIf CI is still running, align polling to check completion boundaries instead of fixed rapid polling.\n\n## Human Gate Rules (Must Ask)\n\nRequire explicit user confirmation for:\n- production/staging deploy beyond agreed scope\n- destructive operations (history rewrite, force push, data-destructive ops)\n- actions with billing/security posture changes\n- secret values not available in repo/runtime\n- ambiguous DoD that materially changes outcome\n\n## Iteration/Stop Conditions\n\nStop and escalate with a concise blocker report when:\n- DoD still fails after max rounds (`2` default)\n- external dependency blocks progress (provider outage, missing creds, account permission)\n- conflicting review instructions cannot both be satisfied\n\nEscalation report must include:\n- what passed\n- what failed\n- evidence (commands/logs/API result)\n- smallest decision needed from user\n\n## Output Contract\n\nWhen claiming completion, always include:\n- acceptance criteria checklist with pass/fail\n- commands/tests run\n- runtime evidence (endpoint/Lambda/log key lines)\n- PR status (new actionable comments count)\n\nDo not claim success without evidence.\n\n## Limitations\n- Use this skill only when the task clearly matches the scope described above.\n- Do not treat the output as a substitute for environment-specific validation, testing, or expert review.\n- Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.","tags":["closed","loop","delivery","antigravity","awesome","skills","sickn33","agent-skills","agentic-skills","ai-agent-skills","ai-agents","ai-coding"],"capabilities":["skill","source-sickn33","skill-closed-loop-delivery","topic-agent-skills","topic-agentic-skills","topic-ai-agent-skills","topic-ai-agents","topic-ai-coding","topic-ai-workflows","topic-antigravity","topic-antigravity-skills","topic-claude-code","topic-claude-code-skills","topic-codex-cli","topic-codex-skills"],"categories":["antigravity-awesome-skills"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/sickn33/antigravity-awesome-skills/closed-loop-delivery","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add sickn33/antigravity-awesome-skills","source_repo":"https://github.com/sickn33/antigravity-awesome-skills","install_from":"skills.sh"}},"qualityScore":"0.700","qualityRationale":"deterministic score 0.70 from registry signals: · indexed on github topic:agent-skills · 34882 github stars · SKILL.md body (3,985 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":"2026-04-24T12:50:46.882Z","embedding":null,"createdAt":"2026-04-18T21:34:29.803Z","updatedAt":"2026-04-24T12:50:46.882Z","lastSeenAt":"2026-04-24T12:50:46.882Z","tsv":"'1':223,341 '10m':357,384 '2':159,248,348,472 '3':258 '3m':343 '4':270 '5':289 '6':308 '6m':350 'a/explanations':120 'accept':15,42,147,161,216,514 'access':133 'account':482 'across':23 'action':281,438,529 'agre':426 'align':402 'allow':199 'alway':512 'ambigu':449 'api/lambda/log':304 'approv':127 'ask':417,571 'avail':446 'avoid':111,331,373 'batch':338,372 'behavior':299 'beyond':425 'billing/security':440 'block':129,476 'blocker':463 'boundari':407,579 'broader':266 'cannot':135,487 'chang':53,251,442,453 'check':95,107,267,317,405 'checklist':516 'checkout':232,235 'ci':398 'claim':510,534 'clarif':573 'classifi':276 'clear':546 'close':2,32 'closed-loop':31 'closed-loop-deliveri':1 'code':8,50,64,88 'coding/fix':77 'collect':344,351,358 'commands/logs/api':500 'commands/tests':519 'comment':91,328,369,381,392,530 'comments/reviews':275,346,362 'complet':12,85,309,406,511 'concis':462 'concret':175 'condit':326,456 'confirm':421 'conflict':484 'continu':200,320 'contract':508 'convert':226 'core':54 'count':531 'creat':187 'create-issue-g':186 'cred':481 'criteria':16,43,148,162,217,230,515,582 'data':435 'data-destruct':434 'decis':310,503 'default':154,158,176,221,336,473 'defin':142,224 'definit':59 'deliv':56 'deliveri':4,34 'delta':345,352 'depend':181,475 'deploy':27,93,103,122,291,294,424 'describ':550 'destruct':428,436 'dev':92,152,247,290,296 'diff':65 'dod':58,149,225,234,307,316,450,466 'done':61,313 'draft':205 'end':82,84 'end-to-end':81 'endpoint':236 'endpoint/lambda/log':523 'environ':151,562 'environment-specif':561 'escal':459,491 'evid':47,305,499,522,537 'exampl':231 'execut':141,183,196,208,220 'expect':80 'expert':567 'explicit':14,125,419 'extern':474 'fail':468,498 'feedback':26 'fetch':273 'final':354 'first':264 'fix':282,410 'focus':262 'forc':432 'gate':180,189,197,414 'give':75 'goal':146,257 'histori':430 'human':126,413 'immedi':374 'implement':24,249 'implementation/deploy/review':209 'includ':494,513 'incomplet':40 'infer':137 'input':139,576 'instead':408 'instruct':486 'intervent':22 'issu':179,188,191,202 'item':284 'iter':156 'iteration/stop':455 'keep':252 'key':524 'like':99 'limit':538 'line':525 'local':260 'loop':3,33,210,272,321 'manual':97 'match':547 'materi':452 'matter':300 'max':155,470 'mere':52 'minim':18,250 'miss':131,164,480,584 'must':10,416,493 'need':269,504 'new':368,528 'noisi':332 'non':280 'non-action':279 'one':371 'op':437 'open':240 'oper':429 'otherwis':319 'outag':479 'outcom':454 'output':507,556 'overview':35 'parti':243 'pass':318,323,496 'pass/fail':518 'payment':244 'permiss':483,577 'point':396 'polici':330 'poll':329,334,377,403,412 'postur':441 'pr':90,108,274,327,526 'prefer':184 'proceed':178,389 'process':366 'prod':121 'production/staging':423 'progress':477 'prompt':98 'propos':173 'provid':172,214,478 'pure':118 'push':433 'q':119 'rapid':411 're':21,106,286,376 're-check':105 're-intervent':20 're-pol':375 're-run':285 'readi':194 'real':303 'remain':360 'repeat':96 'repo/runtime':448 'report':312,464,492 'request':123,165,227 'requir':138,211,418,575 'result':501 'return':237 'review':25,271,485,568 'rewrit':431 'round':157,340,347,355,365,385,471 'rule':55,415 'run':261,287,401,520 'runtim':29,94,292,298,521 'safeti':578 'satisfi':490 'scope':253,427,549 'secret':443 'secrets/account':132 'short':333 'singl':380 'size':66 'skill':72,116,541 'skill-closed-loop-delivery' 'smallest':502 'source-sickn33' 'span':87 'specif':563 'start':219 'status':192,203,527 'still':400,467 'stop':325,386,457,569 'substitut':559 'success':535,581 'target':150 'task':9,38,78,86,128,145,233,256,545 'test':89,101,263,565 'testabl':215,229 'third':242 'third-parti':241 'tight':254 'topic-agent-skills' 'topic-agentic-skills' 'topic-ai-agent-skills' 'topic-ai-agents' 'topic-ai-coding' 'topic-ai-workflows' 'topic-antigravity' 'topic-antigravity-skills' 'topic-claude-code' 'topic-claude-code-skills' 'topic-codex-cli' 'topic-codex-skills' 'treat':36,554 'url':245 'use':5,69,70,114,185,337,539 'user':19,74,169,213,420,506 'user-provid':212 'valid':239,277,283,564 'valu':444 'verif':30,288,293 'verifi':45,259,301 'via':302 'visibl':361,393 'vs':278 'wait':342,349,356,387 'window':339 'without':124,536 'workflow':222","prices":[{"id":"7fd13a1c-7194-4270-b3c1-710a13f82948","listingId":"05b7b51a-5794-4605-8cd8-fc31c98b2f7c","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"sickn33","category":"antigravity-awesome-skills","install_from":"skills.sh"},"createdAt":"2026-04-18T21:34:29.803Z"}],"sources":[{"listingId":"05b7b51a-5794-4605-8cd8-fc31c98b2f7c","source":"github","sourceId":"sickn33/antigravity-awesome-skills/closed-loop-delivery","sourceUrl":"https://github.com/sickn33/antigravity-awesome-skills/tree/main/skills/closed-loop-delivery","isPrimary":false,"firstSeenAt":"2026-04-18T21:34:29.803Z","lastSeenAt":"2026-04-24T12:50:46.882Z"}],"details":{"listingId":"05b7b51a-5794-4605-8cd8-fc31c98b2f7c","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"sickn33","slug":"closed-loop-delivery","github":{"repo":"sickn33/antigravity-awesome-skills","stars":34882,"topics":["agent-skills","agentic-skills","ai-agent-skills","ai-agents","ai-coding","ai-workflows","antigravity","antigravity-skills","claude-code","claude-code-skills","codex-cli","codex-skills","cursor","cursor-skills","developer-tools","gemini-cli","gemini-skills","kiro","mcp","skill-library"],"license":"mit","html_url":"https://github.com/sickn33/antigravity-awesome-skills","pushed_at":"2026-04-24T06:41:17Z","description":"Installable GitHub library of 1,400+ agentic skills for Claude Code, Cursor, Codex CLI, Gemini CLI, Antigravity, and more. Includes installer CLI, bundles, workflows, and official/community skill collections.","skill_md_sha":"6676292c6220d7c0b1c9773280d2d4648a9944af","skill_md_path":"skills/closed-loop-delivery/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/sickn33/antigravity-awesome-skills/tree/main/skills/closed-loop-delivery"},"layout":"multi","source":"github","category":"antigravity-awesome-skills","frontmatter":{"name":"closed-loop-delivery","description":"Use when a coding task must be completed against explicit acceptance criteria with minimal user re-intervention across implementation, review feedback, deployment, and runtime verification."},"skills_sh_url":"https://skills.sh/sickn33/antigravity-awesome-skills/closed-loop-delivery"},"updatedAt":"2026-04-24T12:50:46.882Z"}}