{"id":"9a4c3e93-5b29-41be-bf05-95d61437eeb7","shortId":"MtKGHp","kind":"skill","title":"production-agent-architecture","tagline":"Design production-ready AI agents using a loop-based architecture with explicit tools, context, memory, permissions, observability, evaluation, and rollout planning.","description":"# Production Agent Architecture\n\n## When To Use This Skill\nUse this skill when the task is to:\n- design a new AI agent\n- turn a chatbot idea into a real agent system\n- hand off an agent build to developers or another AI coding agent\n- adapt coding-agent architecture patterns to another domain\n\n## Core Rule\nDo not design the system as a one-shot chatbot unless the user explicitly asks for that.\n\nDefault to a loop-based runtime with:\n1. controller loop\n2. context builder\n3. tool registry\n4. tool executor\n5. session state\n6. permissions and approvals\n7. observability\n8. evaluation\n9. reliability rules\n10. rollout phases\n\n## Design Workflow\n1. define the user goal and success condition\n2. define the controller loop\n3. define 2 to 5 domain tools\n4. define context inputs\n5. define session state\n6. define permissions and approval rules\n7. define observability minimums\n8. define evaluation plan\n9. define reliability and retry rules\n10. define rollout phases\n\n## Required Output\nAlways produce all of the following:\n\n```markdown\n# Agent Build Spec\n\n## Domain\n## User goal\n## Success condition\n## Core workflow\n## Controller loop\n## Context inputs\n## Tools\n## Session state\n## Permissions and approvals\n## Stop conditions\n## Observability\n## Evaluation\n## Reliability and failure handling\n## Rollout phases\n## Future extensions\n```\n\nAfter the spec, also produce:\n- controller loop pseudocode\n- tool contract table with input, output, side effects, timeout, and permission level\n- session state schema\n- approval matrix\n- failure and retry strategy\n- minimum logs and metrics\n\n## Guardrails\n- separate tool schemas from executors\n- treat permissions as runtime policy, not prompt text\n- make approval flows explicit\n- do not pretend a vague idea is production-ready\n- explain the design plainly if the user is learning","tags":["production","agent","architecture","build","your","own","agents","skill","xuanhieu2611","agent-skills","agentic-ai","ai-agents"],"capabilities":["skill","source-xuanhieu2611","skill-production-agent-architecture","topic-agent","topic-agent-skills","topic-agentic-ai","topic-ai-agents","topic-claude-code","topic-codex","topic-cursor","topic-developer","topic-developer-tools","topic-prompt-engineering"],"categories":["build-your-own-agents-skill"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/xuanhieu2611/build-your-own-agents-skill/production-agent-architecture","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add xuanhieu2611/build-your-own-agents-skill","source_repo":"https://github.com/xuanhieu2611/build-your-own-agents-skill","install_from":"skills.sh"}},"qualityScore":"0.460","qualityRationale":"deterministic score 0.46 from registry signals: · indexed on github topic:agent-skills · 20 github stars · SKILL.md body (1,821 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":"2026-04-23T07:01:13.684Z","embedding":null,"createdAt":"2026-04-18T23:06:11.852Z","updatedAt":"2026-04-23T07:01:13.684Z","lastSeenAt":"2026-04-23T07:01:13.684Z","tsv":"'1':107,138 '10':133,186 '2':110,146,153 '3':113,151 '4':116,158 '5':119,155,162 '6':122,166 '7':126,172 '8':128,176 '9':130,180 'adapt':70 'agent':3,10,29,48,56,61,69,73,199 'ai':9,47,67 'also':234 'alway':192 'anoth':66,77 'approv':125,170,218,254,279 'architectur':4,16,30,74 'ask':96 'base':15,104 'build':62,200 'builder':112 'chatbot':51,91 'code':68,72 'coding-ag':71 'condit':145,206,220 'context':20,111,160,211 'contract':240 'control':108,149,209,236 'core':79,207 'default':99 'defin':139,147,152,159,163,167,173,177,181,187 'design':5,44,83,136,294 'develop':64 'domain':78,156,202 'effect':246 'evalu':24,129,178,222 'executor':118,269 'explain':292 'explicit':18,95,281 'extens':230 'failur':225,256 'flow':280 'follow':197 'futur':229 'goal':142,204 'guardrail':264 'hand':58 'handl':226 'idea':52,287 'input':161,212,243 'learn':300 'level':250 'log':261 'loop':14,103,109,150,210,237 'loop-bas':13,102 'make':278 'markdown':198 'matrix':255 'memori':21 'metric':263 'minimum':175,260 'new':46 'observ':23,127,174,221 'one':89 'one-shot':88 'output':191,244 'pattern':75 'permiss':22,123,168,216,249,271 'phase':135,189,228 'plain':295 'plan':27,179 'polici':274 'pretend':284 'produc':193,235 'product':2,7,28,290 'production-agent-architectur':1 'production-readi':6,289 'prompt':276 'pseudocod':238 'readi':8,291 'real':55 'registri':115 'reliabl':131,182,223 'requir':190 'retri':184,258 'rollout':26,134,188,227 'rule':80,132,171,185 'runtim':105,273 'schema':253,267 'separ':265 'session':120,164,214,251 'shot':90 'side':245 'skill':35,38 'skill-production-agent-architecture' 'source-xuanhieu2611' 'spec':201,233 'state':121,165,215,252 'stop':219 'strategi':259 'success':144,205 'system':57,85 'tabl':241 'task':41 'text':277 'timeout':247 'tool':19,114,117,157,213,239,266 'topic-agent' 'topic-agent-skills' 'topic-agentic-ai' 'topic-ai-agents' 'topic-claude-code' 'topic-codex' 'topic-cursor' 'topic-developer' 'topic-developer-tools' 'topic-prompt-engineering' 'treat':270 'turn':49 'unless':92 'use':11,33,36 'user':94,141,203,298 'vagu':286 'workflow':137,208","prices":[{"id":"dc3cd666-a753-4108-86da-ee6b3db83a16","listingId":"9a4c3e93-5b29-41be-bf05-95d61437eeb7","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"xuanhieu2611","category":"build-your-own-agents-skill","install_from":"skills.sh"},"createdAt":"2026-04-18T23:06:11.852Z"}],"sources":[{"listingId":"9a4c3e93-5b29-41be-bf05-95d61437eeb7","source":"github","sourceId":"xuanhieu2611/build-your-own-agents-skill/production-agent-architecture","sourceUrl":"https://github.com/xuanhieu2611/build-your-own-agents-skill/tree/main/skills/production-agent-architecture","isPrimary":false,"firstSeenAt":"2026-04-18T23:06:11.852Z","lastSeenAt":"2026-04-23T07:01:13.684Z"}],"details":{"listingId":"9a4c3e93-5b29-41be-bf05-95d61437eeb7","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"xuanhieu2611","slug":"production-agent-architecture","github":{"repo":"xuanhieu2611/build-your-own-agents-skill","stars":20,"topics":["agent","agent-skills","agentic-ai","ai-agents","claude-code","codex","cursor","developer","developer-tools","prompt-engineering"],"license":"mit","html_url":"https://github.com/xuanhieu2611/build-your-own-agents-skill","pushed_at":"2026-04-03T04:50:57Z","description":"Build AI agents using the same architecture patterns as Claude Code. Skill + 6 examples + runnable Python code. No framework.","skill_md_sha":"5f8eea0f9168ea5cdabd146ee202afb7c68d8318","skill_md_path":"skills/production-agent-architecture/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/xuanhieu2611/build-your-own-agents-skill/tree/main/skills/production-agent-architecture"},"layout":"multi","source":"github","category":"build-your-own-agents-skill","frontmatter":{"name":"production-agent-architecture","description":"Design production-ready AI agents using a loop-based architecture with explicit tools, context, memory, permissions, observability, evaluation, and rollout planning."},"skills_sh_url":"https://skills.sh/xuanhieu2611/build-your-own-agents-skill/production-agent-architecture"},"updatedAt":"2026-04-23T07:01:13.684Z"}}