{"id":"a1ab7e72-e5b4-408f-a3ab-f594792c0470","shortId":"HeJTHe","kind":"skill","title":"greenfield","tagline":"Parallel persona planning for new projects. Research agent runs first to build domain context, then Architect, PM, and Security agents run in parallel. Synthesis agent combines all perspectives into a detailed GSD-style PLAN.md with Tensions section.","description":"# Greenfield Planning Skill\n\n## Trigger\n\nRun once per project: `ws-skills plan`\n\nReads `BRIEF.md` from the project root (or prompts for one). Asks 5 clarifying questions before planning.\n\n## Flow\n\n```\nBrief + Q&A\n    ↓\nResearch agent (sequential)   ← domain landscape, ecosystem, hidden complexity\n    ↓\n┌─────────────────────────────────────┐\n│ Architect │ PM │ Security (parallel)│  ← spawn 3 subagents simultaneously\n└─────────────────────────────────────┘\n    ↓\nSynthesis             ← combines all into PLAN.md\n```\n\n## Agents\n\n### 1. Research (sequential — runs first)\n\nBuilds domain context that all other agents receive. Covers:\n- Existing solutions and their weaknesses\n- Standard and emerging tech stacks for this domain\n- Technologies to avoid and why\n- Non-obvious domain challenges\n- Integration landscape (auth, payments, comms, etc.)\n- Regulatory and compliance context\n- Realistic timeline based on similar projects\n- Hidden complexity — things that take 3x longer than expected\n- Success patterns from the best products in this space\n\nOutput: `research.md`\n\n### 2–4. Architect, PM, Security (parallel subagents)\n\nSpawn all three simultaneously using the Agent tool. Each receives the full brief, Q&A, and research output as context.\n\n```\nAgent 1 — Architect\nAgent 2 — PM           ← launch all three in a single message, do not wait\nAgent 3 — Security\n```\n\nWait for all three to complete before running Synthesis.\n\n**Architect** output: `architect.md`\n- System design overview\n- Tech stack with rationale per layer\n- Module boundaries and interfaces\n- Infrastructure and CI/CD\n- Scaling strategy\n- Technical risks\n\n**PM** output: `pm.md`\n- Phases with tasks and acceptance criteria\n- Success metrics\n- Out of scope items\n- Assumptions\n\n**Security** output: `security.md`\n- Threat model (likelihood + impact)\n- Data classification\n- Auth strategy recommendation\n- Compliance flags\n- Concrete security tasks\n- Urgent flags\n\n### 5. Synthesis\n\nCombines research + all three persona outputs into a single PLAN.md covering:\n- Overview\n- Clarifications table\n- Tech stack\n- Architecture\n- Phases with tasks and acceptance criteria\n- Security plan\n- Success metrics\n- Risks\n- Tensions (unresolved disagreements between personas)\n- Assumptions\n- Out of scope\n- Branch naming (GIT-OS format)\n\nOutput: `PLAN.md`\n\n## Tools\n\n| Action | Tool |\n|--------|------|\n| Read `BRIEF.md` | `Read` |\n| Write persona output files (`architect.md`, `pm.md`, etc.) | `Write` |\n| Spawn Architect, PM, Security personas in parallel | `Agent` (3 calls in one message) |\n| Search the brief for keywords | `Grep` |\n\n## Output Location\n\nAll files written to `.wednesday/plans/` in the target directory:\n\n```\n.wednesday/plans/\n├── research.md    ← domain context\n├── architect.md   ← technical design\n├── pm.md          ← phases and metrics\n├── security.md    ← threat model\n└── PLAN.md        ← combined PRD (primary output)\n```\n\n## Failure Handling\n\nEach agent fails independently. If one fails, the others continue and synthesis runs with whatever data is available. Failed agents show `[partial fallback]` in the progress display.\n\n## Rules\n\n- Branch naming in PLAN.md must follow GIT-OS format\n- Never generate `CODEBASE.md` for greenfield projects — it doesn't exist yet\n- Cost target: under $0.20 per run","tags":["greenfield","agent","skills","wednesday-solutions","agent-skills","ai-agent-skills","ai-skills","claude-code","copilot","cursor","frontend-skills","vibe-coding"],"capabilities":["skill","source-wednesday-solutions","skill-greenfield","topic-agent-skills","topic-ai-agent-skills","topic-ai-skills","topic-claude-code","topic-copilot","topic-cursor","topic-frontend-skills","topic-vibe-coding","topic-vibe-coding-skills"],"categories":["ai-agent-skills"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/wednesday-solutions/ai-agent-skills/greenfield","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add wednesday-solutions/ai-agent-skills","source_repo":"https://github.com/wednesday-solutions/ai-agent-skills","install_from":"skills.sh"}},"qualityScore":"0.524","qualityRationale":"deterministic score 0.52 from registry signals: · indexed on github topic:agent-skills · 148 github stars · SKILL.md body (3,293 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":"2026-05-02T12:54:45.055Z","embedding":null,"createdAt":"2026-04-18T22:09:41.395Z","updatedAt":"2026-05-02T12:54:45.055Z","lastSeenAt":"2026-05-02T12:54:45.055Z","tsv":"'0.20':444 '1':94,195 '2':167,198 '3':85,211,349 '3x':152 '4':168 '5':63,280 'accept':252,303 'action':328 'agent':9,21,26,73,93,105,180,194,197,210,348,393,411 'architect':17,80,169,196,222,342 'architect.md':224,337,375 'architectur':298 'ask':62 'assumpt':260,315 'auth':133,270 'avail':409 'avoid':123 'base':143 'best':160 'boundari':235 'branch':319,420 'brief':69,186,356 'brief.md':53,331 'build':13,99 'call':350 'challeng':130 'ci/cd':240 'clarif':294 'clarifi':64 'classif':269 'codebase.md':432 'combin':27,89,282,386 'comm':135 'complet':218 'complex':79,148 'complianc':139,273 'concret':275 'context':15,101,140,193,374 'continu':401 'cost':441 'cover':107,292 'criteria':253,304 'data':268,407 'design':226,377 'detail':32 'directori':370 'disagr':312 'display':418 'doesn':437 'domain':14,75,100,120,129,373 'ecosystem':77 'emerg':115 'etc':136,339 'exist':108,439 'expect':155 'fail':394,398,410 'failur':390 'fallback':414 'file':336,363 'first':11,98 'flag':274,279 'flow':68 'follow':425 'format':324,429 'full':185 'generat':431 'git':322,427 'git-o':321,426 'greenfield':1,40,434 'grep':359 'gsd':34 'gsd-style':33 'handl':391 'hidden':78,147 'impact':267 'independ':395 'infrastructur':238 'integr':131 'interfac':237 'item':259 'keyword':358 'landscap':76,132 'launch':200 'layer':233 'likelihood':266 'locat':361 'longer':153 'messag':206,353 'metric':255,308,381 'model':265,384 'modul':234 'must':424 'name':320,421 'never':430 'new':6 'non':127 'non-obvi':126 'obvious':128 'one':61,352,397 'os':323,428 'other':400 'output':165,191,223,246,262,287,325,335,360,389 'overview':227,293 'parallel':2,24,83,172,347 'partial':413 'pattern':157 'payment':134 'per':46,232,445 'persona':3,286,314,334,345 'perspect':29 'phase':248,299,379 'plan':4,41,51,67,306 'plan.md':36,92,291,326,385,423 'pm':18,81,170,199,245,343 'pm.md':247,338,378 'prd':387 'primari':388 'product':161 'progress':417 'project':7,47,56,146,435 'prompt':59 'q':70,187 'question':65 'rational':231 'read':52,330,332 'realist':141 'receiv':106,183 'recommend':272 'regulatori':137 'research':8,72,95,190,283 'research.md':166,372 'risk':244,309 'root':57 'rule':419 'run':10,22,44,97,220,404,446 'scale':241 'scope':258,318 'search':354 'section':39 'secur':20,82,171,212,261,276,305,344 'security.md':263,382 'sequenti':74,96 'show':412 'similar':145 'simultan':87,177 'singl':205,290 'skill':42,50 'skill-greenfield' 'solut':109 'source-wednesday-solutions' 'space':164 'spawn':84,174,341 'stack':117,229,297 'standard':113 'strategi':242,271 'style':35 'subag':86,173 'success':156,254,307 'synthesi':25,88,221,281,403 'system':225 'tabl':295 'take':151 'target':369,442 'task':250,277,301 'tech':116,228,296 'technic':243,376 'technolog':121 'tension':38,310 'thing':149 'threat':264,383 'three':176,202,216,285 'timelin':142 'tool':181,327,329 'topic-agent-skills' 'topic-ai-agent-skills' 'topic-ai-skills' 'topic-claude-code' 'topic-copilot' 'topic-cursor' 'topic-frontend-skills' 'topic-vibe-coding' 'topic-vibe-coding-skills' 'trigger':43 'unresolv':311 'urgent':278 'use':178 'wait':209,213 'weak':112 'wednesday/plans':366,371 'whatev':406 'write':333,340 'written':364 'ws':49 'ws-skill':48 'yet':440","prices":[{"id":"3f528a3e-7921-465f-ae47-e57d2cfcc1d4","listingId":"a1ab7e72-e5b4-408f-a3ab-f594792c0470","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"wednesday-solutions","category":"ai-agent-skills","install_from":"skills.sh"},"createdAt":"2026-04-18T22:09:41.395Z"}],"sources":[{"listingId":"a1ab7e72-e5b4-408f-a3ab-f594792c0470","source":"github","sourceId":"wednesday-solutions/ai-agent-skills/greenfield","sourceUrl":"https://github.com/wednesday-solutions/ai-agent-skills/tree/main/skills/greenfield","isPrimary":false,"firstSeenAt":"2026-04-18T22:09:41.395Z","lastSeenAt":"2026-05-02T12:54:45.055Z"}],"details":{"listingId":"a1ab7e72-e5b4-408f-a3ab-f594792c0470","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"wednesday-solutions","slug":"greenfield","github":{"repo":"wednesday-solutions/ai-agent-skills","stars":148,"topics":["agent-skills","ai-agent-skills","ai-skills","claude-code","copilot","cursor","frontend-skills","product-engineering-skills","vibe-coding","vibe-coding-skills"],"license":"mit","html_url":"https://github.com/wednesday-solutions/ai-agent-skills","pushed_at":"2026-04-24T12:57:49Z","description":"Pre-configured agent skills for Vibe Coded projects. These skills provide AI coding assistants (Claude Code, Cursor, etc.) with specific guidelines for code quality and design standards.","skill_md_sha":"16d176fc6bf304645a559be20fd7843b3328fbbe","skill_md_path":"skills/greenfield/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/wednesday-solutions/ai-agent-skills/tree/main/skills/greenfield"},"layout":"multi","source":"github","category":"ai-agent-skills","frontmatter":{"name":"greenfield","license":"MIT","description":"Parallel persona planning for new projects. Research agent runs first to build domain context, then Architect, PM, and Security agents run in parallel. Synthesis agent combines all perspectives into a detailed GSD-style PLAN.md with Tensions section."},"skills_sh_url":"https://skills.sh/wednesday-solutions/ai-agent-skills/greenfield"},"updatedAt":"2026-05-02T12:54:45.055Z"}}