[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M72YW2AXSX8VK7TZQC3BDF4Q":3,"skill-same:01M72YW2AXSX8VK7TZQC3BDF4Q":97},{"skillNo":4,"name":5,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":6,"repositoryNo":11,"starsCount":12,"licenseSpdxId":13,"licenseFileUrl":6,"sourceHtmlUrl":14,"categories":15,"keywords":16,"publishedAt":17,"linkOnly":18,"markdownContent":19,"markdownBody":19,"frontMatter":20,"refName":21,"commitSha":22,"skillDirectory":23,"forksCount":24,"repositoryHtmlUrl":25,"repositoryHomepageUrl":26,"ownerHtmlUrl":27,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":28,"install":29,"resources":42},"01M72YW2AXSX8VK7TZQC3BDF4Q","dev-builder",null,"zxpmail\u002FReqForge","zxpmail","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F45926210?v=4","01M72YNG60707F44824AGAD6HM","01M72YNG6GD4BTF9Q8V35FQMYY",18,"MIT","https:\u002F\u002Fgithub.com\u002Fzxpmail\u002FReqForge\u002Fblob\u002Fmain\u002Fcore\u002Fskills\u002Fdev-builder\u002FSKILL.md",[],[],"2026-11-11 15:11:14",false,"\u003C!-- forge: dev-builder v1.2 -->\n---\nname: dev-builder\ndescription: Used when DEV-PLAN.md is ready and the user says to start coding or continue developing the next Phase. Sets up the skeleton for new projects, implements features by Phase for existing projects.\nversion: 1.2.0\nupdated: 2026-06-27\nrequires: []\n---\n\n\u003C!-- begin: task -->\n[Task]\n    **Initialization Mode (0-to-1)**: No code + has DEV-PLAN.md -> set up project skeleton according to tech stack, install dependencies, configure development environment, complete Phase 1.\n    **Phase 1 的核心不是\"写功能\"，是\"放骨架\"**:\n      - 先写领域模型、核心类型定义、数据流接口\n      - 再写 Validator 模式、错误处理公约、中间件链\n      - 这些就是项目独有的\"天理\"——牛的骨节结构\n      - 后续所有 Phase 的代码都是从这些骨架生长出来的\n      - 所以 Phase 1 宁可慢一点把骨架放稳，也不要为了\"快点看到功能\"而跳过\n      - 骨架放好后，后续 Phase 的代码模型自然续写——因为\"这个项目的代码长什么样\"已经定了\n\n    **Continuous Development Mode**: Has code + has DEV-PLAN.md -> develop by Phase, **one Phase per \u002Fdev-builder invocation**. Each Phase: Plan Mode to plan implementation -> per-Task review + commit -> Phase four-step verification -> user confirmation -> **force stop** (suggests running **\u002Fcode-review** for a holistic Phase-level review). User must call \u002Fdev-builder again for next Phase.\n\n    **Change-Scoped Mode**: Invoked from `\u002Fchange-manager apply` with `change-name=\u003Cname>` -> read `changes\u002F\u003Cname>\u002F` (specs, design, tasks), execute **only** unchecked items in `changes\u002F\u003Cname>\u002Ftasks.md`. Do not pull unrelated DEV-PLAN Phases. Still uses implementer + TDD + two-tier review per Task.\n\n\u003C!-- end: task -->\n\u003C!-- begin: not-for -->\n[Not For]\n    - Fixing bugs in existing code -> use \u002Fbug-fixer instead\n    - Reviewing code quality -> use \u002Fcode-review instead\n    - Planning development phases -> use \u002Fdev-planner instead\n    - Gathering requirements -> use \u002Fproduct-spec-builder instead\n\n\u003C!-- end: not-for -->\n\u003C!-- begin: dependency-check -->\n[Dependency Check]\n    Executed automatically as the first step when the Skill starts.\n\n    Required:\n    - Product-Spec.md -> if missing, prompt user to call \u002Fproduct-spec-builder first\n    - DEV-PLAN.md -> if missing, prompt user to call \u002Fdev-planner first\n    - All system tools and runtime environments listed in the DEV-PLAN tech stack table\n\n    Optional:\n    - Design-Brief.md -> if missing, mark as \"no design specification mode\"\n    - DESIGN.md -> if present, use frozen tokens for Tailwind\u002Ftheme (priority over Brief for exact values)\n    - Design tool MCP -> if missing, mark as \"no design draft mode\"\n    - gh CLI -> if available, can automatically create GitHub repo and push\n    - playwright -> if available, can do UI automated testing\n    - **Dependency Graph** (`dep-graph`) -> if available, enables blast-radius analysis for impact assessment and risk-scored complexity gating\n\n    Installation Strategy:\n    - When required dependencies are missing or version requirements not met, the Agent autonomously determines the installation method and installs directly — no manual user operation needed\n    - If user permissions or interaction is needed, prompt the user to act\n    - When optional dependencies are missing, mark as degraded mode and continue working — do not block the workflow\n\n\u003C!-- end: dependency-check -->\n\u003C!-- begin: first-principles -->\n[First Principles]\n    **编码前必读** `references\u002Ffirst-principles.md`。TDD、implementer 隔离、验证即证据 — 非协商。\n\n\u003C!-- end: first-principles -->\n\u003C!-- begin: shared-discipline -->\n[Shared Discipline]\n    Karpathy 四原则 → `..\u002F_shared\u002Fkarpathy-discipline.md`（全文 `core\u002Fdocs\u002Fbehavior-rules.md`）\n    finding 的 `action` 三分（auto-fix\u002Fask-user\u002Fno-op）驱动 Step 14.6 立即 escalate → `..\u002F_shared\u002Ffinding-actions.md`\n\n\u003C!-- end: shared-discipline -->\n\u003C!-- begin: hard-gate -->\n[HARD-GATE]\n    主 Session **MUST NOT** 直接 Write\u002FEdit 业务代码；每 Task dispatch implementer 并创建 `.forge\u002Fimplementer-session.json`；首次改代码前 **MUST** git worktree。\n\n    **Exception 1 — UI \u002F Integration Phase**: 主 session 可直接 Write\u002FEdit，不强制 implementer 和 worktree。理由：UI 代码需要完整组件上下文，implementer 隔离得不偿失（Dogfood #2 验证）。\n\n    **Exception 2 — Small Phase (any Nature)**: 当 Phase 的 Key Files ≤3 且 Deliverables ≤5（即\"Small Phase\"），主 session 可直接 Write\u002FEdit，不强制 implementer 和 worktree。理由：小型 Phase 的 implementer 冷启动开销 > 实际编码工作量（Dogfood #3 验证：Greenfield CLI 全部 5 Phase 均小到不值得隔离）。尺寸判定详见 `references\u002Fworkflow.md` § Nature Gate Step 1.5。\n\n    以上豁免均仅限当前 Phase，下一 Phase 重新评估。\n\n    机器门与 Session 生命周期 → `..\u002F_shared\u002Fhard-gate-summary.md`（Hook 拦截为准）\n\n\u003C!-- end: hard-gate -->\n\u003C!-- begin: output-style -->\n[Output Style]\n    → `..\u002F_shared\u002Foutput-style-concise.md`\n    → Phase 完成必须附加 `..\u002F_shared\u002Foutput-status-protocol.md`（Status: DONE \u002F DONE_WITH_CONCERNS \u002F BLOCKED \u002F NEEDS_CONTEXT）\n\n\u003C!-- end: output-style -->\n\u003C!-- begin: file-structure -->\n[File Structure]\n    ```\n    dev-builder\u002F\n    ├── SKILL.md                    # 入口（本文件）\n    ├── commands\u002Fdev-builder.md     # 命令摘要\n    └── references\u002F\n        ├── first-principles.md     # 编码原则（必读）\n        ├── workflow.md             # 完整 Workflow（按需）\n        ├── development-rules-checklist.md\n        ├── development-strategies.md\n        ├── anti-rationalization.md\n        ├── anti-ai-slop-checklist.md\n        ├── sub-agent-isolation.md\n        ├── phase-completion-assessment.md\n        └── zoom-out-pass.md\n    ..\u002F_shared\u002F                     # 跨 Skill 共享指针\n    ```\n\n\u003C!-- end: file-structure -->\n\u003C!-- begin: output-artifacts -->\n[Output Artifacts]\n    - **Project code** — Complete project code under the \\\u003Cproject-name\\>\u002F directory\n    - **Git commits** — Atomic commits (phase-N: \u002F fix: \u002F feat: \u002F refactor: \u002F chore:)\n    - **..\u002F..\u002F.needs-review** — Review status indicator (clear or needs_review)\n    - **memory\u002Ftask-history.md** — Always append after Task completion (mandatory)\n    - **memory\u002Fdecisions-log.md** — Append when a technical decision was made during the Task\n    - **memory\u002Fproject-memory.md** — Update when architecture facts or constraints change\n\n\u003C!-- end: output-artifacts -->\n\u003C!-- begin: development-rules-checklist -->\n[Development Rules Checklist]\n    **执行前读取** `references\u002Fdevelopment-rules-checklist.md`\n\n\u003C!-- end: development-rules-checklist -->\n\u003C!-- begin: development-strategies -->\n[Development Strategies]\n    **按需读取** `references\u002Fdevelopment-strategies.md`\n\n\u003C!-- end: development-strategies -->\n\u003C!-- begin: gotchas -->\n[Gotchas]\n    **Plan-not-loaded**: Starting implementation without reading the current DEV-PLAN.md Phase → building the wrong thing. Always read DEV-PLAN.md first, confirm the Phase and Task, then code.\n    **Skipping Environment-First**: Jumping into feature code before the project skeleton compiles and runs. No code on a broken foundation. The first task of any Phase should be making things runnable.\n    **Phase scope creep**: \"I'll just add this small improvement while I'm coding\" → that's how Phases inflate and never finish. One Phase, one goal. Additional improvements go to the feedback channel or next Phase. Before adding scope, check DEV-PLAN **Scope amendment criteria** — no qualifying user evidence, no build.\n    **Editing Spec\u002FPlan during build**: Patching Product-Spec.md or DEV-PLAN.md to excuse implementation drift violates the prepare.py boundary. Route scope changes through change-manager or replan.\n    **Missing verification**: Completing a Task without compile\u002Ffunc\u002Fregression verification. Every Task must have its own mini-verification before Phase Assessment.\n    **Difficulty-blind execution**: Every Phase has a **Difficulty** level. 🔴 高 = 怵然为戒，每段代码追加自我评审，动刀甚微；🟢 低 = 快速通过，标准流程即可。不按难度调整行为，就像庖丁遇到筋骨交错还一刀砍过去。执行前先读 Phase 的 Difficulty 字段。\n    **Nature-blind execution**: Every Phase also has a **Nature** field (Backend\u002FUI\u002FData\u002FIntegration). Backend\u002FData → dispatch implementer; UI\u002FIntegration → main session writes directly. 读 Nature 失败意味着做错 dispatch 决策——UI Phase 走了 implementer 浪费上下文中断开销，Backend Phase 走了主 session 污染上下文。执行前同时读 Difficulty + Nature。\n    **Skipping Nature Gate**: Launching into Step 2 per-Task loop without checking `references\u002Fworkflow.md` § Nature Gate → incorrectly defaults to implementer for UI Phases. Always read Nature before first Task dispatch.\n    **Skipping 善刀而藏之**: 做完一个 Phase 不收刀——不在 Spec 追加新发现的难点、不记决策日志、不清理上下文。下一个 Phase 会在上一轮的噪声上启动。杀完牛不等于结束，收刀才是。\n\n\u003C!-- end: gotchas -->\n\u003C!-- begin: anti-rationalization-checklist -->\n[Anti-Rationalization Checklist]\n    **遇阻力时读取** `references\u002Fanti-rationalization.md`\n\n\u003C!-- end: anti-rationalization-checklist -->\n\u003C!-- begin: phase-completion-assessment -->\n[Phase Completion Assessment]\n    **Phase 结束 Step 3 必须按此文执行** `references\u002Fphase-completion-assessment.md`\n\n\u003C!-- end: phase-completion-assessment -->\n\u003C!-- begin: development-dimension-checklist -->\n[Dimension Checklist]\n    See [references\u002Fdevelopment-dimension-checklist.md](references\u002Fdevelopment-dimension-checklist.md) for the full dimension checklist.\n\n    Must-have dimensions:\n    - **Code Structure & Modularity**: consistent module layout, clear layer separation\n    - **Type Safety**: strict static typing, boundary validation\n    - **Error Handling**: categorized errors, async handlers, readable user-facing messages\n    - **Testing Coverage**: unit + integration + e2e covering all branches\n    - **Security Basics**: input validation, auth, CSRF\u002FXSS protection\n    - **API Design Consistency**: uniform endpoints, correct status codes, versioning\n\n\u003C!-- end: development-dimension-checklist -->\n\u003C!-- begin: quality-rubric -->\n[Quality Rubric]\n    10-item, 20-point scoring system. Ship threshold: **≥ 16** with no critical item scoring 0.\n\n    | # | Dimension | Pts | Critical | Scoring |\n    |---|-----------|-----|----------|---------|\n    | 1 | Spec alignment | 2 | YES | 2 = All implemented behavior matches Product-Spec.md; 1 = Minor deviation in non-core path; 0 = Implementation contradicts spec |\n    | 2 | Test coverage | 2 | YES | 2 = Unit + integration + e2e covering all branches and edge cases; 1 = Core paths covered but gaps in error\u002Fedge paths; 0 = No tests or critical paths untested |\n    | 3 | Code structure | 2 | — | 2 = Modular, clear layer separation, consistent file organization; 1 = Some tangling or inconsistent naming; 0 = Monolithic or chaotic structure |\n    | 4 | Type safety | 2 | — | 2 = Strict typing throughout, no `any`, no unsafe casts; 1 = Occasional `any` or loose types in non-critical areas; 0 = Widespread `any` or type errors |\n    | 5 | Error handling | 2 | — | 2 = Proper error boundaries, categorized errors, clear user-facing messages; 1 = Async errors caught but no boundary\u002Fcategory; 0 = Uncaught rejections or silent failures |\n    | 6 | State management | 2 | — | 2 = Correct data flow, immutable patterns, predictable mutation; 1 = Minor anti-patterns but functional; 0 = Mutations cause stale\u002Fincorrect state |\n    | 7 | Performance awareness | 2 | — | 2 = Query optimized, bundle size managed, render efficiency considered; 1 = Some inefficiencies but not blocking; 0 = N+1 queries, no memo, large bundles |\n    | 8 | Security basics | 2 | YES | 2 = Input validation, auth checks, XSS\u002FCSRF prevention in place; 1 = Validation present but incomplete; 0 = No security measures, injection vectors open |\n    | 9 | Git discipline | 2 | — | 2 = Frequent atomic commits, meaningful messages, logical grouping; 1 = Commits bundled or vague messages; 0 = Single mega-commit or no versioned intermediate state |\n    | 10 | Phase completion completeness | 2 | YES | 2 = All checklist items addressed, no TODO\u002FFIXME left; 1 = Minor items skipped but documented; 0 = Key deliverables missing or incomplete |\n    | 11 | Code minimality | 2 | — | 2 = No dead code, no speculative abstraction, no unused imports\u002Fdeps, prefers well-maintained small packages over reinventing wheels, avoids pulling heavy deps for trivial use; 1 = Minor over-engineering (\u003C 10% reducible); 0 = Unnecessary abstraction layers, bloated deps, dead code left behind |\n\n    **Scoring**: Run `pnpm validate-skill --score core\u002Fskills\u002Fdev-builder` to compute.\n\u003C!-- end: quality-rubric -->\n\u003C!-- begin: workflow -->\n[Workflow]\n    1. Run [Dependency Check]\n    2. Read `references\u002Ffirst-principles.md`\n    3. Route via [Initialization] → **必须先 Read `references\u002Fworkflow.md` 对应章节**，再进入 Phase Execution（未读禁止写业务代码）\n       - **Initialization Mode** — greenfield scaffold + Phase 1\n       - **Continuous Development Mode** — Loading → Phase Execution → Verification → Force Stop\n       - **Change-Scoped Mode** — same Continuous flow; tasks from `changes\u002F\u003Cname>\u002Ftasks.md` only\n    4. Phase 交付前执行 `references\u002Fanti-ai-slop-checklist.md` 自检\n    5. Optional read-only: `references\u002Fzoom-out-pass.md` when user asks zoom out\n    6. YOLO (`FORGE_MODE=yolo`) overrides → see `references\u002Fworkflow.md` § YOLO Mode\n\n\u003C!-- end: workflow -->\n\u003C!-- begin: initialization -->\n[Initialization]\n    Detect project state, route to the corresponding mode:\n    - No code + has DEV-PLAN.md -> Initialization Mode\n    - Has code + has DEV-PLAN.md -> Continuous Development Mode\n    - No DEV-PLAN.md -> prompt to call \u002Fdev-planner first\n    - No Product-Spec.md -> prompt to call \u002Fproduct-spec-builder first\n\n\u003C!-- end: initialization -->\n",{},"main","bf75b50757318dc73f1fee44c5a56f6324242411","core\u002Fskills\u002Fdev-builder",5,"https:\u002F\u002Fgithub.com\u002Fzxpmail\u002FReqForge","https:\u002F\u002Fgithub.com\u002Fzxpmail\u002FReqForge#readme","https:\u002F\u002Fgithub.com\u002Fzxpmail","2026-10-07 09:22:30",{"source":7,"skillName":5,"commands":30,"prompt":40,"zipAvailable":41},[31,34,37],{"tool":32,"command":33},"npx","npx skills add zxpmail\u002FReqForge --skill dev-builder",{"tool":35,"command":36},"bunx","bunx skills add zxpmail\u002FReqForge --skill dev-builder",{"tool":38,"command":39},"pnpm","pnpm dlx skills add zxpmail\u002FReqForge --skill dev-builder","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fzxpmail\u002FReqForge\u002Fblob\u002Fmain\u002Fcore\u002Fskills\u002Fdev-builder\u002FSKILL.md",true,[43,49,53,57,61,65,69,73,77,81,85,89,93],{"path":44,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":47,"sourceHtmlUrl":48},"commands\u002Fdev-builder.md","REFERENCE","参考文档",678,"https:\u002F\u002Fgithub.com\u002Fzxpmail\u002FReqForge\u002Fblob\u002Fmain\u002Fcore\u002Fskills\u002Fdev-builder\u002Fcommands\u002Fdev-builder.md",{"path":50,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":51,"sourceHtmlUrl":52},"references\u002Fanti-ai-slop-checklist.md",4766,"https:\u002F\u002Fgithub.com\u002Fzxpmail\u002FReqForge\u002Fblob\u002Fmain\u002Fcore\u002Fskills\u002Fdev-builder\u002Freferences\u002Fanti-ai-slop-checklist.md",{"path":54,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":55,"sourceHtmlUrl":56},"references\u002Fanti-rationalization.md",3051,"https:\u002F\u002Fgithub.com\u002Fzxpmail\u002FReqForge\u002Fblob\u002Fmain\u002Fcore\u002Fskills\u002Fdev-builder\u002Freferences\u002Fanti-rationalization.md",{"path":58,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":59,"sourceHtmlUrl":60},"references\u002Fdevelopment-dimension-checklist.md",4120,"https:\u002F\u002Fgithub.com\u002Fzxpmail\u002FReqForge\u002Fblob\u002Fmain\u002Fcore\u002Fskills\u002Fdev-builder\u002Freferences\u002Fdevelopment-dimension-checklist.md",{"path":62,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":63,"sourceHtmlUrl":64},"references\u002Fdevelopment-rules-checklist.md",10418,"https:\u002F\u002Fgithub.com\u002Fzxpmail\u002FReqForge\u002Fblob\u002Fmain\u002Fcore\u002Fskills\u002Fdev-builder\u002Freferences\u002Fdevelopment-rules-checklist.md",{"path":66,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":67,"sourceHtmlUrl":68},"references\u002Fdevelopment-strategies.md",5528,"https:\u002F\u002Fgithub.com\u002Fzxpmail\u002FReqForge\u002Fblob\u002Fmain\u002Fcore\u002Fskills\u002Fdev-builder\u002Freferences\u002Fdevelopment-strategies.md",{"path":70,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":71,"sourceHtmlUrl":72},"references\u002Ffirst-principles.md",6404,"https:\u002F\u002Fgithub.com\u002Fzxpmail\u002FReqForge\u002Fblob\u002Fmain\u002Fcore\u002Fskills\u002Fdev-builder\u002Freferences\u002Ffirst-principles.md",{"path":74,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":75,"sourceHtmlUrl":76},"references\u002Fgc-audit-routing.md",2492,"https:\u002F\u002Fgithub.com\u002Fzxpmail\u002FReqForge\u002Fblob\u002Fmain\u002Fcore\u002Fskills\u002Fdev-builder\u002Freferences\u002Fgc-audit-routing.md",{"path":78,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":79,"sourceHtmlUrl":80},"references\u002Fphase-completion-assessment.md",10164,"https:\u002F\u002Fgithub.com\u002Fzxpmail\u002FReqForge\u002Fblob\u002Fmain\u002Fcore\u002Fskills\u002Fdev-builder\u002Freferences\u002Fphase-completion-assessment.md",{"path":82,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":83,"sourceHtmlUrl":84},"references\u002Fsub-agent-isolation.md",4985,"https:\u002F\u002Fgithub.com\u002Fzxpmail\u002FReqForge\u002Fblob\u002Fmain\u002Fcore\u002Fskills\u002Fdev-builder\u002Freferences\u002Fsub-agent-isolation.md",{"path":86,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":87,"sourceHtmlUrl":88},"references\u002Fworkflow.md",32468,"https:\u002F\u002Fgithub.com\u002Fzxpmail\u002FReqForge\u002Fblob\u002Fmain\u002Fcore\u002Fskills\u002Fdev-builder\u002Freferences\u002Fworkflow.md",{"path":90,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":91,"sourceHtmlUrl":92},"references\u002Fzoom-out-pass.md",1033,"https:\u002F\u002Fgithub.com\u002Fzxpmail\u002FReqForge\u002Fblob\u002Fmain\u002Fcore\u002Fskills\u002Fdev-builder\u002Freferences\u002Fzoom-out-pass.md",{"path":94,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":95,"sourceHtmlUrl":96},"skill.json",926,"https:\u002F\u002Fgithub.com\u002Fzxpmail\u002FReqForge\u002Fblob\u002Fmain\u002Fcore\u002Fskills\u002Fdev-builder\u002Fskill.json",[98,105,112,118,124,130],{"skillNo":99,"name":100,"nameZh":6,"description":101,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":6,"repositoryNo":11,"starsCount":12,"licenseSpdxId":13,"licenseFileUrl":6,"sourceHtmlUrl":102,"categories":103,"keywords":104,"publishedAt":17},"01M72YWW227N72XBY2DB439PZ9","feedback-writer","Analyze and record user corrections, feedback, and behavior preferences as structured feedback files for the evolution engine.","https:\u002F\u002Fgithub.com\u002Fzxpmail\u002FReqForge\u002Fblob\u002Fmain\u002Ftest-demo\u002F.claude\u002Fskills\u002Ffeedback-writer\u002FSKILL.md",[],[],{"skillNo":106,"name":107,"nameZh":6,"description":108,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":6,"repositoryNo":11,"starsCount":12,"licenseSpdxId":13,"licenseFileUrl":6,"sourceHtmlUrl":109,"categories":110,"keywords":111,"publishedAt":17},"01M72YWTKT8GWESEYTJXE6HXG4","evolution-engine","Scan accumulated feedback to identify patterns, generate evolution proposals for SKILL.md, rules, or new Skills.","https:\u002F\u002Fgithub.com\u002Fzxpmail\u002FReqForge\u002Fblob\u002Fmain\u002Ftest-demo\u002F.claude\u002Fskills\u002Fevolution-engine\u002FSKILL.md",[],[],{"skillNo":113,"name":114,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":6,"repositoryNo":11,"starsCount":12,"licenseSpdxId":13,"licenseFileUrl":6,"sourceHtmlUrl":115,"categories":116,"keywords":117,"publishedAt":17},"01M72YWP1C81B4NNENKME8NPMF","db-migration-checker","https:\u002F\u002Fgithub.com\u002Fzxpmail\u002FReqForge\u002Fblob\u002Fmain\u002Feval-output\u002Fnew-skill-created\u002Fskills\u002Fdb-migration-checker\u002FSKILL.md",[],[],{"skillNo":119,"name":120,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":6,"repositoryNo":11,"starsCount":12,"licenseSpdxId":13,"licenseFileUrl":6,"sourceHtmlUrl":121,"categories":122,"keywords":123,"publishedAt":17},"01M72YWN2GJN9XMS4D0BAPDZJR","skill-builder","https:\u002F\u002Fgithub.com\u002Fzxpmail\u002FReqForge\u002Fblob\u002Fmain\u002Fcore\u002Fskills\u002Fskill-builder\u002FSKILL.md",[],[],{"skillNo":125,"name":126,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":6,"repositoryNo":11,"starsCount":12,"licenseSpdxId":13,"licenseFileUrl":6,"sourceHtmlUrl":127,"categories":128,"keywords":129,"publishedAt":17},"01M72YWK6PBEEV8P1EFYX5TGBD","request-dispatcher","https:\u002F\u002Fgithub.com\u002Fzxpmail\u002FReqForge\u002Fblob\u002Fmain\u002Fcore\u002Fskills\u002Frequest-dispatcher\u002FSKILL.md",[],[],{"skillNo":131,"name":132,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":6,"repositoryNo":11,"starsCount":12,"licenseSpdxId":13,"licenseFileUrl":6,"sourceHtmlUrl":133,"categories":134,"keywords":135,"publishedAt":17},"01M72YWG2KHA4PRCPSSD9J7166","reqforge-greenkeeper","https:\u002F\u002Fgithub.com\u002Fzxpmail\u002FReqForge\u002Fblob\u002Fmain\u002Fcore\u002Fskills\u002Freqforge-greenkeeper\u002FSKILL.md",[],[]]