{"id":"f403c8a7-9f65-46e6-98e4-9b49e9d893bf","shortId":"eRTrqp","kind":"skill","title":"bmad-architect","tagline":"Architecture skill for BMAD. Use for bmad:architecture and bmad:gate-check to produce system design and verify requirement coverage.","description":"# BMAD Architect\n\n## Trigger Intents\n\n- `bmad:architecture`\n- `bmad:gate-check`\n\n## Workflow Variants\n\n1. `architecture`\n- Produce architecture decisions, components, interfaces, and data model.\n\n2. `gate-check`\n- Validate architecture quality and requirement coverage before implementation.\n\nMode selection:\n- Run `architecture` when `docs/bmad/architecture.md` does not exist or requirements changed materially.\n- Run `gate-check` when architecture exists and the team needs a formal implementation-readiness decision.\n\n## Inputs\n\n- planning artifact (`docs/bmad/prd.md` or `docs/bmad/tech-spec.md`)\n- architecture artifact (`docs/bmad/architecture.md`) for gate-check mode\n- `bmad/project.yaml` level and constraints\n- known integration, compliance, and scalability constraints\n\n## Language Guard (Mandatory)\n\nEnforce language selection separately for chat responses and generated artifacts.\n\nChat language (`communication_language`) fallback order:\n\n1. `language.communication_language` from `bmad/project.yaml`\n2. `English`\n\nRules for chat responses:\n\n- Use the resolved chat language for all assistant responses (questions, status updates, summaries, and handoff notes).\n- Do not switch chat language unless the user explicitly requests a different language in the current thread.\n\nArtifact language (`document_output_language`) fallback order:\n\n1. `language.document_output_language` from `bmad/project.yaml`\n2. `English`\n\nRules for generated artifacts:\n\n- Use the resolved artifact language for all generated BMAD documents and structured artifacts.\n- write prose and field values in the resolved document language\n- avoid mixed-language requirement clauses with English modal verbs (for example, `System shall` followed by non-English text)\n- allow English acronyms/abbreviations in non-English sentences (for example, `API`, `SLA`, `KPI`, `OAuth`, `WCAG`)\n- Keep code snippets, CLI commands, file paths, and identifiers in their original technical form.\n\n## Mandatory Reference Load\n\nBefore executing `architecture` or `gate-check`, read `REFERENCE.md` first.\nTreat `REFERENCE.md` as required context, then load relevant resources and templates.\n\n## Output Contract\n\n- `architecture` -> `docs/bmad/architecture.md`\n- `gate-check` -> `docs/bmad/gate-check.md`\n\n## Core Workflow\n\n1. Define system boundaries and architecture drivers.\n2. Select architecture pattern and justify trade-offs.\n3. Design components, APIs, and data model.\n4. Map NFRs to concrete controls and design choices.\n5. Run gate-check using criteria from `resources/gate-check-criteria.md` and report pass/fail gaps.\n\n## Gate-Check Criteria\n\nLoad and apply `resources/gate-check-criteria.md` as the single source of truth for:\n- coverage metric formulas\n- blocker classification rules\n- `PASS` / `CONDITIONAL PASS` / `FAIL` thresholds\n\n## Gate-Check Artifact Format\n\nAlways save `docs/bmad/gate-check.md` using `templates/gate-check.template.md`.\n\nRequired sections:\n- Executive Summary with decision (`PASS`, `CONDITIONAL PASS`, or `FAIL`)\n- Requirements Coverage (FR and NFR totals, coverage percentages, missing items)\n- Architecture Quality Assessment (checklist score and failed checks)\n- Issues (Blockers, Major Concerns, Minor Issues)\n- Recommendations\n- Gate Decision (threshold comparison and rationale)\n- Next Steps\n- Appendix with detailed FR/NFR mapping evidence\n\n## Script Selection\n\n- `architecture` mode:\n  ```bash\n  bash scripts/nfr-checklist.sh\n  bash scripts/validate-architecture.sh docs/bmad/architecture.md\n  ```\n- `gate-check` mode:\n  ```bash\n  bash scripts/validate-architecture.sh docs/bmad/architecture.md\n  bash scripts/nfr-checklist.sh\n  ```\n\n## Template Map\n\n- `templates/architecture.template.md`\n- Why: full architecture structure with design decisions and NFR mapping.\n\n- `templates/gate-check.template.md`\n- Why: deterministic gate-check report format with objective pass criteria.\n\n## Reference Map\n\n- `REFERENCE.md`\n- Must read first for architecture workflow details and design decision quality criteria.\n\n- `resources/architecture-patterns.md`\n- Use when selecting monolith, modular monolith, microservices, or other patterns.\n\n- `resources/nfr-mapping.md`\n- Use to map performance, security, scalability, and reliability requirements.\n\n- `resources/gate-check-criteria.md`\n- Use as mandatory criteria source for gate-check scoring and decision thresholds.\n\n## Quality Gates\n\n- architecture decisions trace back to requirements\n- interfaces and data model are explicit\n- NFR coverage is documented, not implied\n- critical risks and mitigations are listed\n- implementation can proceed without unresolved blockers","tags":["bmad","architect","codex","skills","xmm","agent-skills","agentic-ai","agentic-workflow","agile","ai-coding-agent","ai-skills","analysis"],"capabilities":["skill","source-xmm","skill-bmad-architect","topic-agent-skills","topic-agentic-ai","topic-agentic-workflow","topic-agile","topic-ai-coding-agent","topic-ai-skills","topic-analysis","topic-bmad","topic-code-reviews","topic-codex","topic-codex-cli","topic-developer-tool"],"categories":["codex-bmad-skills"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/xmm/codex-bmad-skills/bmad-architect","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add xmm/codex-bmad-skills","source_repo":"https://github.com/xmm/codex-bmad-skills","install_from":"skills.sh"}},"qualityScore":"0.465","qualityRationale":"deterministic score 0.47 from registry signals: · indexed on github topic:agent-skills · 30 github stars · SKILL.md body (4,624 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":"2026-05-01T07:01:06.613Z","embedding":null,"createdAt":"2026-04-18T22:22:53.237Z","updatedAt":"2026-05-01T07:01:06.613Z","lastSeenAt":"2026-05-01T07:01:06.613Z","tsv":"'1':37,132,183,301 '2':47,137,189,308 '3':317 '4':324 '5':333 'acronyms/abbreviations':240 'allow':238 'alway':377 'api':248,320 'appendix':426 'appli':352 'architect':3,26 'architectur':4,11,30,38,40,52,62,77,95,272,293,306,310,403,434,457,484,529 'artifact':91,96,125,176,194,198,207,375 'assess':405 'assist':150 'avoid':218 'back':532 'bash':436,437,439,446,447,450 'blocker':364,412,558 'bmad':2,7,10,13,25,29,31,203 'bmad-architect':1 'bmad/project.yaml':103,136,188 'boundari':304 'chang':70 'chat':121,126,141,146,162 'check':16,34,50,75,101,276,297,337,348,374,410,444,470,522 'checklist':406 'choic':332 'classif':365 'claus':223 'cli':256 'code':254 'command':257 'communic':128 'comparison':421 'complianc':109 'compon':42,319 'concern':414 'concret':328 'condit':368,389 'constraint':106,112 'context':284 'contract':292 'control':329 'core':299 'coverag':24,56,361,394,399,542 'criteria':339,349,476,491,517 'critic':547 'current':174 'data':45,322,537 'decis':41,88,387,419,461,489,525,530 'defin':302 'design':20,318,331,460,488 'detail':428,486 'determinist':467 'differ':170 'docs/bmad/architecture.md':64,97,294,441,449 'docs/bmad/gate-check.md':298,379 'docs/bmad/prd.md':92 'docs/bmad/tech-spec.md':94 'document':178,204,216,544 'driver':307 'enforc':116 'english':138,190,225,236,239,244 'evid':431 'exampl':229,247 'execut':271,384 'exist':67,78 'explicit':167,540 'fail':370,392,409 'fallback':130,181 'field':211 'file':258 'first':279,482 'follow':232 'form':266 'formal':84 'format':376,472 'formula':363 'fr':395 'fr/nfr':429 'full':456 'gap':345 'gate':15,33,49,74,100,275,296,336,347,373,418,443,469,521,528 'gate-check':14,32,48,73,99,274,295,335,346,372,442,468,520 'generat':124,193,202 'guard':114 'handoff':157 'identifi':261 'implement':58,86,553 'implementation-readi':85 'impli':546 'input':89 'integr':108 'intent':28 'interfac':43,535 'issu':411,416 'item':402 'justifi':313 'keep':253 'known':107 'kpi':250 'languag':113,117,127,129,134,147,163,171,177,180,186,199,217,221 'language.communication':133 'language.document':184 'level':104 'list':552 'load':269,286,350 'major':413 'mandatori':115,267,516 'map':325,430,453,464,478,506 'materi':71 'metric':362 'microservic':499 'minor':415 'miss':401 'mitig':550 'mix':220 'mixed-languag':219 'modal':226 'mode':59,102,435,445 'model':46,323,538 'modular':497 'monolith':496,498 'must':480 'need':82 'next':424 'nfr':397,463,541 'nfrs':326 'non':235,243 'non-english':234,242 'note':158 'oauth':251 'object':474 'off':316 'order':131,182 'origin':264 'output':179,185,291 'pass':367,369,388,390,475 'pass/fail':344 'path':259 'pattern':311,502 'percentag':400 'perform':507 'plan':90 'proceed':555 'produc':18,39 'prose':209 'qualiti':53,404,490,527 'question':152 'rational':423 'read':277,481 'readi':87 'recommend':417 'refer':268,477 'reference.md':278,281,479 'relev':287 'reliabl':511 'report':343,471 'request':168 'requir':23,55,69,222,283,382,393,512,534 'resolv':145,197,215 'resourc':288 'resources/architecture-patterns.md':492 'resources/gate-check-criteria.md':341,353,513 'resources/nfr-mapping.md':503 'respons':122,142,151 'risk':548 'rule':139,191,366 'run':61,72,334 'save':378 'scalabl':111,509 'score':407,523 'script':432 'scripts/nfr-checklist.sh':438,451 'scripts/validate-architecture.sh':440,448 'section':383 'secur':508 'select':60,118,309,433,495 'sentenc':245 'separ':119 'shall':231 'singl':356 'skill':5 'skill-bmad-architect' 'sla':249 'snippet':255 'sourc':357,518 'source-xmm' 'status':153 'step':425 'structur':206,458 'summari':155,385 'switch':161 'system':19,230,303 'team':81 'technic':265 'templat':290,452 'templates/architecture.template.md':454 'templates/gate-check.template.md':381,465 'text':237 'thread':175 'threshold':371,420,526 'topic-agent-skills' 'topic-agentic-ai' 'topic-agentic-workflow' 'topic-agile' 'topic-ai-coding-agent' 'topic-ai-skills' 'topic-analysis' 'topic-bmad' 'topic-code-reviews' 'topic-codex' 'topic-codex-cli' 'topic-developer-tool' 'total':398 'trace':531 'trade':315 'trade-off':314 'treat':280 'trigger':27 'truth':359 'unless':164 'unresolv':557 'updat':154 'use':8,143,195,338,380,493,504,514 'user':166 'valid':51 'valu':212 'variant':36 'verb':227 'verifi':22 'wcag':252 'without':556 'workflow':35,300,485 'write':208","prices":[{"id":"8be96180-5fb8-4610-b348-6371f146d8e5","listingId":"f403c8a7-9f65-46e6-98e4-9b49e9d893bf","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"xmm","category":"codex-bmad-skills","install_from":"skills.sh"},"createdAt":"2026-04-18T22:22:53.237Z"}],"sources":[{"listingId":"f403c8a7-9f65-46e6-98e4-9b49e9d893bf","source":"github","sourceId":"xmm/codex-bmad-skills/bmad-architect","sourceUrl":"https://github.com/xmm/codex-bmad-skills/tree/main/skills/bmad-architect","isPrimary":false,"firstSeenAt":"2026-04-18T22:22:53.237Z","lastSeenAt":"2026-05-01T07:01:06.613Z"}],"details":{"listingId":"f403c8a7-9f65-46e6-98e4-9b49e9d893bf","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"xmm","slug":"bmad-architect","github":{"repo":"xmm/codex-bmad-skills","stars":30,"topics":["agent-skills","agentic-ai","agentic-workflow","agile","ai-coding-agent","ai-skills","analysis","bmad","code-reviews","codex","codex-cli","developer-tool","openai-codex","product-management","prompt-engineering","software-architecture","workflow"],"license":"other","html_url":"https://github.com/xmm/codex-bmad-skills","pushed_at":"2026-04-20T10:20:30Z","description":"BMAD skills and workflows for OpenAI Codex (App, CLI, Web): intent-based execution, YAML project state, and reusable skill packs for planning, architecture, sprint delivery, development, and code review.","skill_md_sha":"c797d22d396772c640fb65d551d922a5f22665b3","skill_md_path":"skills/bmad-architect/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/xmm/codex-bmad-skills/tree/main/skills/bmad-architect"},"layout":"multi","source":"github","category":"codex-bmad-skills","frontmatter":{"name":"bmad-architect","description":"Architecture skill for BMAD. Use for bmad:architecture and bmad:gate-check to produce system design and verify requirement coverage."},"skills_sh_url":"https://skills.sh/xmm/codex-bmad-skills/bmad-architect"},"updatedAt":"2026-05-01T07:01:06.613Z"}}