[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6XDEZ78G56PBPP7ZYV9PMTF":3,"skill-same:01M6XDEZ78G56PBPP7ZYV9PMTF":77},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":23,"publishedAt":43,"linkOnly":44,"markdownContent":45,"markdownBody":46,"frontMatter":47,"refName":50,"commitSha":51,"skillDirectory":52,"forksCount":14,"repositoryHtmlUrl":53,"repositoryHomepageUrl":6,"ownerHtmlUrl":54,"ownerBlogUrl":6,"ownerDescription":55,"updatedAt":56,"install":57,"resources":70},"01M6XDEZ78G56PBPP7ZYV9PMTF","spec",null,"Create structured engineering specs through interactive pairing. Use when the user wants to create a spec, design doc, technical specification, architecture document, RFC, or engineering plan. Triggers on \"let's spec this out\", \"create a spec\", \"design doc\", \"write a technical plan\", \"plan the architecture\". Works for greenfield and brownfield projects. Outputs to docs\u002Fspecs\u002F.","kaiohenricunha\u002Fdotbabel","kaiohenricunha","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F69804490?v=4","01M6XCZD37W0KPCDGEA40Y610V","Kaio Cunha","01M6XCZD7SX714D114H3ZPNW09",0,"MIT","https:\u002F\u002Fgithub.com\u002Fkaiohenricunha\u002Fdotbabel\u002Fblob\u002Fmain\u002Fplugins\u002Fdotbabel\u002Ftemplates\u002Fclaude\u002Fskills\u002Fspec\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000015","视觉设计","SCENARIO","使用场景",[5,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42],"when","design","outputs","pairing","for","technical specification","triggers","brownfield","specification","document","use","write","technical","design doc","projects","plan","doc","create","architecture","2026-11-10 10:12:30",false,"---\n# dotbabel:generated projection=sha256:973f2e15a256a65051a1d8fc910157e5f7a5138985d26870a4441b7df8b87fb9 source=skills\u002Fspec\u002FSKILL.md — do not edit directly\nargument-hint: \"[spec-name] [description] [--brownfield]\"\ndescription: |\n  Create structured engineering specs through interactive pairing. Use when the user wants to create a spec, design doc, technical specification, architecture document, RFC, or engineering plan. Triggers on \"let's spec this out\", \"create a spec\", \"design doc\", \"write a technical plan\", \"plan the architecture\". Works for greenfield and brownfield projects. Outputs to docs\u002Fspecs\u002F.\nname: spec\n---\n\n# Spec-Driven Design\n\nCreate structured engineering specifications through interactive, phased collaboration.\nOutputs organized markdown files into `docs\u002Fspecs\u002F\u003Cspec-name>\u002F` of the current project.\n\n## Arguments\n\n- `$0` — spec name (kebab-case). If not provided, ask the user.\n- `$1` — brief description of what the spec covers (free text, optional). Used to seed the README subtitle and give context for Phase 2 questions. If not provided, proceed normally — the user will supply context interactively.\n- `--brownfield` flag — if present, include a `current-state\u002F` folder for existing system analysis.\n\n## Workflow Overview\n\nThere are 3 phases. **Always start with Phase 1.** Do not skip ahead.\n\n---\n\n### Phase 1: Scaffold\n\nCreate the spec directory and empty section files. This gives the user a concrete structure to react to.\n\n**Steps:**\n\n1. Determine the spec name from `$0` or ask the user. Capture the description from `$1` if provided.\n2. Create the directory structure:\n\n```\ndocs\u002Fspecs\u002F\u003Cspec-name>\u002F\n├── spec.json\n├── README.md\n├── spec\u002F\n│   ├── 1-problem-motivation.md\n│   ├── 2-scope.md\n│   ├── 3-high-level-architecture.md\n│   ├── 4-data-flow-components.md\n│   ├── 5-interfaces-apis.md\n│   ├── 6-implementation-plan.md\n│   ├── 7-non-functional-requirements.md\n│   └── 8-risks-alternatives.md\n├── research\u002F\n│   └── sources.md\n└── current-state\u002F              ← only if --brownfield\n    └── analysis.md\n```\n\n3. Each spec section file gets the scaffold template (see \"Section Scaffolds\" below).\n4. `spec.json` gets the scaffold below. **This file is not optional** —\n   `dotbabel-validate-specs` rejects a spec directory without one, so a scaffold\n   that omits it produces a spec that fails validation on its first run.\n5. The README gets a status dashboard showing all sections as `[ ] empty`.\n6. Tell the user the scaffold is created.\n   - **If `$1` was provided:** \"Scaffold created. Based on your description — _{$1}_ — let's start with §1 Problem\u002FMotivation. What specifically is broken or missing that this addresses?\"\n   - **If no description:** \"Scaffold created. Let's fill it in — starting with §1 Problem\u002FMotivation. What problem are we solving?\"\n\n**Do not fill in any content during scaffolding.** The scaffold is purely structural (the description goes only in the README header, not into section bodies).\n\n---\n\n### Phase 2: Interactive Fill-In\n\nWalk through each section with the user. For each section:\n\n1. **Ask** a focused question about what belongs in that section.\n2. **Listen** — only write what the user provides or explicitly approves. Never invent requirements.\n3. **Write** — update the section file with the user's input, properly formatted.\n4. **Tag** — assign constraint IDs where appropriate (ARCH-1, PERF-1, etc.) for cross-referencing.\n5. **Update README** — mark the section status as `[~] in-progress` or `[x] done`.\n6. **Offer next** — suggest moving to the next section, or ask if they want to go deeper.\n\n**Section order is suggested, not mandatory.** If the user wants to jump to §6 Implementation Plan, go there. If they want to revisit §2 Scope after filling §5, do that.\n\n**Rules:**\n\n- Only add content the user explicitly provides or approves.\n- Tag constraints with IDs (e.g., PERF-1, SEC-3, ARCH-1) for cross-referencing between sections.\n- Place constraints in the section they **shape**, not just where they're mentioned.\n- Cross-reference between sections: \"See ARCH-1 in §3\" or \"Feeds into §6 Phase 2\".\n- Empty sections are fine — mark them and move on. The user fills them when ready.\n\n#### Research Sources\n\nAs the user references documents, codebases, or external resources:\n\n1. Assign each a **DOC-N** identifier.\n2. Add it to `research\u002Fsources.md` with a one-line description.\n3. Tag which spec sections it feeds (e.g., \"Feeds: §3, §5, §7\").\n\n#### Deep Dives\n\nFor complex sections (especially §4 Data Flow and §6 Implementation Plan), offer to generate\n**analysis prompts** the user can run in separate Claude Code sessions. Use the templates\nfrom [references\u002Fcc-prompt-templates.md](references\u002Fcc-prompt-templates.md).\n\nWhen the user pastes back results, integrate findings into the appropriate section files.\n\n---\n\n### Phase 3: Finalize\n\nWhen all sections are filled (or the user says \"done\" \u002F \"finalize\"):\n\n1. Do a **structural** consistency pass — cross-references valid, constraint IDs used, no orphaned references.\n2. Do a **content** consistency pass — three checks, each reported to the user:\n   - Every quality claim in §1\u002F§2 has a matching constraint in §7.\n   - Every measurement in `research\u002F` (DOC-N) became a §7 constraint, or was dropped with a stated reason.\n   - §6.4 covers every output §5 ships.\n3. Update README with final status: all sections marked `[x] done` or `[ ] skipped`.\n4. Add a \"Quick Start\" section to README pointing to the most important sections.\n5. Tell the user the spec is complete and where to find it.\n\n---\n\n## Section Scaffolds\n\nUse these templates when creating each section file in Phase 1.\n\n### 1-problem-motivation.md\n\n```markdown\n# §1 — Problem \u002F Motivation\n\n> Why does this exist? What's broken? Why now?\n\n## The Question\n\n\u003C!-- The single question a user asks that this feature answers.\n     One sentence, in the user's words, not the system's.\n     If you cannot write it, the spec is not ready for §2. -->\n\n## Why\n\n\u003C!-- What is the core problem or opportunity? -->\n\n## What\n\n\u003C!-- What does the solution look like at a high level? -->\n\n## Why Now\n\n\u003C!-- What changed that makes this urgent? Incidents? Scale? Business needs? -->\n```\n\n### 2-scope.md\n\n```markdown\n# §2 — Scope\n\n> What's in, what's out, and where are the boundaries?\n\n## In Scope\n\n\u003C!-- Bulleted list of what this spec covers -->\n\n## Out of Scope\n\n\u003C!-- Bulleted list of what this spec explicitly does NOT cover, with brief reasoning -->\n\n## Boundaries\n\n\u003C!-- What files\u002Fsystems\u002Fservices does this touch vs. not touch? -->\n\n| Touches | Does Not Touch |\n| ------- | -------------- |\n|         |                |\n\n## Urgency\n\n\u003C!-- How time-sensitive is this? Any hard deadlines? -->\n```\n\n### 3-high-level-architecture.md\n\n```markdown\n# §3 — High-Level Architecture\n\n> System view: components, data stores, external dependencies, deployment.\n\n## System Overview\n\n\u003C!-- How does the system fit together at the highest level? -->\n\n## Data Stores\n\n\u003C!-- What databases, caches, queues, or storage systems are involved? -->\n\n| Store | Role | Access Pattern |\n| ----- | ---- | -------------- |\n|       |      |                |\n\n## External APIs \u002F Dependencies\n\n\u003C!-- What external services does this depend on? -->\n\n| Service | Purpose | Rate Limits \u002F Constraints |\n| ------- | ------- | ------------------------- |\n|         |         |                           |\n\n## Deployment\n\n\u003C!-- Where does this run? Regions, runtimes, infrastructure. -->\n```\n\n### 4-data-flow-components.md\n\n```markdown\n# §4 — Data Flow \u002F Components\n\n> Current state analysis + target architecture.\n\n## Current State\n\n\u003C!-- How does data flow through the system today? -->\n\n## Component Boundaries\n\n\u003C!-- What are the key modules\u002Fservices and their responsibilities? -->\n\n## Shared State\n\n\u003C!-- What state is shared between components? How is it synchronized? -->\n\n## Target Architecture\n\n\u003C!-- What does the new architecture look like? Key decisions and rationale. -->\n\n### Key Decisions\n\n\u003C!-- Tag each as KD-N for cross-referencing -->\n```\n\n### 5-interfaces-apis.md\n\n```markdown\n# §5 — Interfaces and APIs\n\n> External APIs, internal endpoints, database schemas.\n\n## External APIs\n\n\u003C!-- APIs consumed from third parties -->\n\n## Internal APIs\n\n\u003C!-- Endpoints this system exposes -->\n\n## Database Schema\n\n\u003C!-- Key tables\u002Fcollections and their structure -->\n```\n\n### 6-implementation-plan.md\n\n```markdown\n# §6 — Implementation Plan\n\n> Phases, workstreams, prompts, tests, migrations, rollback.\n\n## 6.1 Phased Rollout\n\n\u003C!-- Build order, dependencies, what's parallelizable -->\n\n## 6.2 Workstream Breakdown\n\n\u003C!-- Parallel work tracks, interface contracts between them -->\n\n## 6.3 Prompt Sequence\n\n\u003C!-- One prompt per implementation unit -->\n\u003C!--\nEach prompt should include:\n- \u003Cread-first> block listing 4-6 specific source files\n- The existing \u002Fplan workflow; recommend an explicit reasoning effort only when\n  the active runtime supports that control. Do not invent a slash command.\n- TDD test names (the test is the first deliverable)\n- Exact file paths to create\u002Fmodify\n- A \u003Cproof> block: the tests, the command that runs the new behavior, what the\n  run must show, and the done_when list (references\u002Fcc-prompt-templates.md).\n  Each unit is one \u002Fdeliver contract.\n-->\n\n## 6.4 Testing Strategy\n\n\u003C!--\nFor each kind below: name where it applies, or write \"N\u002FA — \u003Creason>\".\nAn unexplained blank is not an answer.\n\n  · unit            — per-function behavior\n  · contract        — payload\u002FAPI invariants that must never silently break\n  · property        — invariants across the input space, not per-case\n  · golden\u002Ffixture  — locks behavior against a validated reference\n  · integration     — real dependencies; replay against historical data\n  · mutation        — proves the suite constrains the code; use wherever the\n                      code and its tests share an author, human or agent\n  · statistical     — REQUIRED when the output is probabilistic, ranked, or\n                      scored: name the metric, the baseline, and the sample size\n  · load\u002Ftorture    — behavior at full scale\n  · post-deploy     — what reality reports back after shipping\n  · QA (run it)     — REQUIRED per unit: the command, page, or call that runs\n                      the new behavior, and what it must show; \"N\u002FA\" only when\n                      the unit has no runnable surface\n-->\n\n| Unit | Kinds applied | QA (run it) | N\u002FA + reason |\n| ---- | ------------- | ----------- | ------------ |\n|      |               |             |              |\n\n## 6.5 Migration Sequence\n\n\u003C!-- Numbered steps, all additive (no breaking changes), deployable before code -->\n\n## 6.6 Rollback Plan\n\n\u003C!-- Scenario → action table, coexistence strategy for old\u002Fnew -->\n\n| Scenario | Action | Notes |\n| -------- | ------ | ----- |\n|          |        |       |\n```\n\n### 7-non-functional-requirements.md\n\n```markdown\n# §7 — Non-Functional Requirements\n\n> Performance, reliability, operational, security constraints.\n\n\u003C!-- A threshold constraint names its metric, its value, and what happens on breach.\n     \"Fast\", \"reliable\", \"a drift alarm\" are adjectives, not constraints.\n     An invariant needs no number — \"all writes are atomic\", \"never overwrites a\n     user file\" are binary and testable exactly as written. Reach for a number\n     only where you are promising a threshold.\n     A measurement recorded in research\u002F is not a constraint until it lands here.\n     `dotbabel-validate-specs` enforces this: a comparative with no value fails. -->\n\n## Performance\n\n\u003C!-- Tag each as PERF-N -->\n\n## Reliability\n\n\u003C!-- Tag each as REL-N -->\n\n## Operational\n\n\u003C!-- Tag each as OPS-N -->\n\n## Security\n\n\u003C!-- Tag each as SEC-N -->\n```\n\n### 8-risks-alternatives.md\n\n```markdown\n# §8 — Risks and Alternatives\n\n> Known risks with mitigations, rejected approaches with reasoning.\n\n## Risks\n\n\u003C!-- Tag each as R-N -->\n\n| ID  | Risk | Likelihood | Impact | Mitigation |\n| --- | ---- | ---------- | ------ | ---------- |\n|     |      |            |        |            |\n\n## Rejected Alternatives\n\n\u003C!-- Tag each as A-N. Include WHY it was rejected — future readers need the reasoning. -->\n```\n\n### spec.json\n\nThe machine-readable half of the spec. `dotbabel-validate-specs` requires\n`id`, `title`, `status`, `owners`, `linked_paths` and `acceptance_commands`;\n`acceptance_criteria` is optional but is what makes the spec's claims\nmachine-checkable rather than prose.\n\n```json\n{\n  \"id\": \"\u003Cspec-name>\",\n  \"title\": \"\u003CSpec Title>\",\n  \"status\": \"draft\",\n  \"owners\": [\"@\u003Cowner>\"],\n  \"linked_paths\": [\"\u003Cpath\u002Fthis\u002Fspec\u002Fgoverns>\"],\n  \"acceptance_commands\": [\"\u003Cthe command that proves this spec works>\"],\n  \"depends_on_specs\": [],\n  \"active_prs\": [],\n  \"acceptance_criteria\": [\n    {\n      \"id\": \"AC-1\",\n      \"status\": \"planned\",\n      \"given\": \"\u003Cthe precondition, in the system's own terms>\",\n      \"when\": \"\u003Cthe action taken>\",\n      \"then\": \"\u003Cthe observable outcome — one claim, not a list>\",\n      \"tests\": [\n        {\n          \"file\": \"\u003Cpath\u002Fto\u002Fthe.test.file>\",\n          \"name\": \"\u003Cthe exact test name, copied verbatim from the test>\"\n        }\n      ],\n      \"argv\": [\n        \"\u003Cthe command that runs this test>\",\n        \"\u003Cits arguments>\",\n        \"\u003Cpath\u002Fto\u002Fthe.test.file>\",\n        \"--reporter=junit\",\n        \"--outputFile=.dotbabel\u002Fcriteria\u002FAC-1.junit.xml\"\n      ],\n      \"report\": { \"format\": \"junit-xml\", \"path\": \".dotbabel\u002Fcriteria\u002FAC-1.junit.xml\" }\n    }\n  ]\n}\n```\n\n**Every new criterion starts `planned`.** A criterion becomes `active` only in\nthe pull request that adds the tests it names — seeding `active` at scaffold\ntime asserts, from the first commit, that tests exist which do not, and the\nverification command will fail looking for them.\n\n`linked_paths` and `acceptance_commands` must both be **non-empty** — the\nvalidator rejects an empty array for either, so replace the placeholders rather\nthan deleting them. `depends_on_specs` and `active_prs` must be present but may\nstay empty.\n\nKeep `linked_paths` narrow. A broad entry (`package.json`, a whole `src\u002F`)\ndrags unrelated changes into this spec's merge gate, so add a path only when a\nchange there genuinely needs this spec's criteria to pass.\n\nLeave `tests[].name` and `argv` as placeholders until the tests exist. The name\nmust later match the test **verbatim** — verification looks it up by exact\nstring in the JUnit report, so a paraphrase silently never matches.\n\n`argv` is runner-agnostic: the only fixed requirement is that the command emits\nJUnit XML at `report.path`, which every major runner can do. For a Node project\nthat is `[\"npx\", \"vitest\", \"run\", \"\u003Cfile>\", \"--reporter=junit\",\n\"--outputFile=…\"]`; Go uses `gotestsum --junitfile`, Python `pytest\n--junitxml`. Write the one your project actually runs rather than copying the\nNode form.\n\n### README.md\n\n```markdown\n# {Spec Name} — Engineering Spec\n\n> {description from $1, or omit this line if none provided}\n>\n> Created: {date}\n\n## Status\n\n| #   | Section                     | Status    |\n| --- | --------------------------- | --------- |\n| 1   | Problem \u002F Motivation        | [ ] empty |\n| 2   | Scope                       | [ ] empty |\n| 3   | High-Level Architecture     | [ ] empty |\n| 4   | Data Flow \u002F Components      | [ ] empty |\n| 5   | Interfaces and APIs         | [ ] empty |\n| 6   | Implementation Plan         | [ ] empty |\n| 7   | Non-Functional Requirements | [ ] empty |\n| 8   | Risks and Alternatives      | [ ] empty |\n\n## Quick Start\n\n\u003C!-- Filled in during finalization -->\n\n## Research Sources\n\nSee [research\u002Fsources.md](research\u002Fsources.md) for indexed source documents.\n```\n\n### research\u002Fsources.md\n\n```markdown\n# Research Sources\n\n> Indexed documents feeding into this spec. Each tagged with which sections it informs.\n\n\u003C!-- Format:\n- **DOC-N**: {title} — {one-line description}. Feeds: §N, §N.\n-->\n```\n\n### current-state\u002Fanalysis.md (brownfield only)\n\n```markdown\n# Current State Analysis\n\n> Analysis of the existing system being redesigned.\n\n## System Overview\n\n\u003C!-- What exists today? -->\n\n## Pain Points\n\n\u003C!-- What's broken, slow, or hard to maintain? -->\n\n## Preserved Behaviors\n\n\u003C!-- What MUST remain the same after the rewrite? -->\n```\n\n---\n\n## Constraint ID Conventions\n\n| Prefix | Domain         | Lives In            |\n| ------ | -------------- | ------------------- |\n| ARCH-N | Architecture   | §3, §4              |\n| IMPL-N | Implementation | §6                  |\n| TEST-N | Testing        | §6                  |\n| PERF-N | Performance    | §7                  |\n| REL-N  | Reliability    | §7                  |\n| OPS-N  | Operational    | §7                  |\n| SEC-N  | Security       | §7                  |\n| KD-N   | Key Decisions  | §4 target arch      |\n| R-N    | Risks          | §8                  |\n| A-N    | Alternatives   | §8                  |\n| DOC-N  | Research docs  | research\u002Fsources.md |\n\n---\n\n## Key Principles\n\n1. **User drives content, Claude drives structure.** Never invent requirements. Place user input in the correct section with proper tagging.\n2. **Scaffold first, fill interactively.** Create the empty structure immediately so the user can see what they're building toward.\n3. **Cross-reference everything.** Constraints reference sections. Docs reference sections. Prompts reference docs and files.\n4. **TDD-first in §6.** Every prompt in the Implementation Plan must list failing test names before implementation instructions.\n5. **Empty sections are fine.** Don't fill sections preemptively. Mark them empty and move on.\n6. **Constraints belong where they shape design.** A testing requirement goes in §6, not §7. A modularity requirement goes in §4, not §7. Operational limits go in §7.\n7. **One question at a time.** Don't overwhelm the user with multiple questions. Ask about one section or one aspect, then write what they give you.\n","# Spec-Driven Design\n\nCreate structured engineering specifications through interactive, phased collaboration.\nOutputs organized markdown files into `docs\u002Fspecs\u002F\u003Cspec-name>\u002F` of the current project.\n\n## Arguments\n\n- `$0` — spec name (kebab-case). If not provided, ask the user.\n- `$1` — brief description of what the spec covers (free text, optional). Used to seed the README subtitle and give context for Phase 2 questions. If not provided, proceed normally — the user will supply context interactively.\n- `--brownfield` flag — if present, include a `current-state\u002F` folder for existing system analysis.\n\n## Workflow Overview\n\nThere are 3 phases. **Always start with Phase 1.** Do not skip ahead.\n\n---\n\n### Phase 1: Scaffold\n\nCreate the spec directory and empty section files. This gives the user a concrete structure to react to.\n\n**Steps:**\n\n1. Determine the spec name from `$0` or ask the user. Capture the description from `$1` if provided.\n2. Create the directory structure:\n\n```\ndocs\u002Fspecs\u002F\u003Cspec-name>\u002F\n├── spec.json\n├── README.md\n├── spec\u002F\n│   ├── 1-problem-motivation.md\n│   ├── 2-scope.md\n│   ├── 3-high-level-architecture.md\n│   ├── 4-data-flow-components.md\n│   ├── 5-interfaces-apis.md\n│   ├── 6-implementation-plan.md\n│   ├── 7-non-functional-requirements.md\n│   └── 8-risks-alternatives.md\n├── research\u002F\n│   └── sources.md\n└── current-state\u002F              ← only if --brownfield\n    └── analysis.md\n```\n\n3. Each spec section file gets the scaffold template (see \"Section Scaffolds\" below).\n4. `spec.json` gets the scaffold below. **This file is not optional** —\n   `dotbabel-validate-specs` rejects a spec directory without one, so a scaffold\n   that omits it produces a spec that fails validation on its first run.\n5. The README gets a status dashboard showing all sections as `[ ] empty`.\n6. Tell the user the scaffold is created.\n   - **If `$1` was provided:** \"Scaffold created. Based on your description — _{$1}_ — let's start with §1 Problem\u002FMotivation. What specifically is broken or missing that this addresses?\"\n   - **If no description:** \"Scaffold created. Let's fill it in — starting with §1 Problem\u002FMotivation. What problem are we solving?\"\n\n**Do not fill in any content during scaffolding.** The scaffold is purely structural (the description goes only in the README header, not into section bodies).\n\n---\n\n### Phase 2: Interactive Fill-In\n\nWalk through each section with the user. For each section:\n\n1. **Ask** a focused question about what belongs in that section.\n2. **Listen** — only write what the user provides or explicitly approves. Never invent requirements.\n3. **Write** — update the section file with the user's input, properly formatted.\n4. **Tag** — assign constraint IDs where appropriate (ARCH-1, PERF-1, etc.) for cross-referencing.\n5. **Update README** — mark the section status as `[~] in-progress` or `[x] done`.\n6. **Offer next** — suggest moving to the next section, or ask if they want to go deeper.\n\n**Section order is suggested, not mandatory.** If the user wants to jump to §6 Implementation Plan, go there. If they want to revisit §2 Scope after filling §5, do that.\n\n**Rules:**\n\n- Only add content the user explicitly provides or approves.\n- Tag constraints with IDs (e.g., PERF-1, SEC-3, ARCH-1) for cross-referencing between sections.\n- Place constraints in the section they **shape**, not just where they're mentioned.\n- Cross-reference between sections: \"See ARCH-1 in §3\" or \"Feeds into §6 Phase 2\".\n- Empty sections are fine — mark them and move on. The user fills them when ready.\n\n#### Research Sources\n\nAs the user references documents, codebases, or external resources:\n\n1. Assign each a **DOC-N** identifier.\n2. Add it to `research\u002Fsources.md` with a one-line description.\n3. Tag which spec sections it feeds (e.g., \"Feeds: §3, §5, §7\").\n\n#### Deep Dives\n\nFor complex sections (especially §4 Data Flow and §6 Implementation Plan), offer to generate\n**analysis prompts** the user can run in separate Claude Code sessions. Use the templates\nfrom [references\u002Fcc-prompt-templates.md](references\u002Fcc-prompt-templates.md).\n\nWhen the user pastes back results, integrate findings into the appropriate section files.\n\n---\n\n### Phase 3: Finalize\n\nWhen all sections are filled (or the user says \"done\" \u002F \"finalize\"):\n\n1. Do a **structural** consistency pass — cross-references valid, constraint IDs used, no orphaned references.\n2. Do a **content** consistency pass — three checks, each reported to the user:\n   - Every quality claim in §1\u002F§2 has a matching constraint in §7.\n   - Every measurement in `research\u002F` (DOC-N) became a §7 constraint, or was dropped with a stated reason.\n   - §6.4 covers every output §5 ships.\n3. Update README with final status: all sections marked `[x] done` or `[ ] skipped`.\n4. Add a \"Quick Start\" section to README pointing to the most important sections.\n5. Tell the user the spec is complete and where to find it.\n\n---\n\n## Section Scaffolds\n\nUse these templates when creating each section file in Phase 1.\n\n### 1-problem-motivation.md\n\n```markdown\n# §1 — Problem \u002F Motivation\n\n> Why does this exist? What's broken? Why now?\n\n## The Question\n\n\u003C!-- The single question a user asks that this feature answers.\n     One sentence, in the user's words, not the system's.\n     If you cannot write it, the spec is not ready for §2. -->\n\n## Why\n\n\u003C!-- What is the core problem or opportunity? -->\n\n## What\n\n\u003C!-- What does the solution look like at a high level? -->\n\n## Why Now\n\n\u003C!-- What changed that makes this urgent? Incidents? Scale? Business needs? -->\n```\n\n### 2-scope.md\n\n```markdown\n# §2 — Scope\n\n> What's in, what's out, and where are the boundaries?\n\n## In Scope\n\n\u003C!-- Bulleted list of what this spec covers -->\n\n## Out of Scope\n\n\u003C!-- Bulleted list of what this spec explicitly does NOT cover, with brief reasoning -->\n\n## Boundaries\n\n\u003C!-- What files\u002Fsystems\u002Fservices does this touch vs. not touch? -->\n\n| Touches | Does Not Touch |\n| ------- | -------------- |\n|         |                |\n\n## Urgency\n\n\u003C!-- How time-sensitive is this? Any hard deadlines? -->\n```\n\n### 3-high-level-architecture.md\n\n```markdown\n# §3 — High-Level Architecture\n\n> System view: components, data stores, external dependencies, deployment.\n\n## System Overview\n\n\u003C!-- How does the system fit together at the highest level? -->\n\n## Data Stores\n\n\u003C!-- What databases, caches, queues, or storage systems are involved? -->\n\n| Store | Role | Access Pattern |\n| ----- | ---- | -------------- |\n|       |      |                |\n\n## External APIs \u002F Dependencies\n\n\u003C!-- What external services does this depend on? -->\n\n| Service | Purpose | Rate Limits \u002F Constraints |\n| ------- | ------- | ------------------------- |\n|         |         |                           |\n\n## Deployment\n\n\u003C!-- Where does this run? Regions, runtimes, infrastructure. -->\n```\n\n### 4-data-flow-components.md\n\n```markdown\n# §4 — Data Flow \u002F Components\n\n> Current state analysis + target architecture.\n\n## Current State\n\n\u003C!-- How does data flow through the system today? -->\n\n## Component Boundaries\n\n\u003C!-- What are the key modules\u002Fservices and their responsibilities? -->\n\n## Shared State\n\n\u003C!-- What state is shared between components? How is it synchronized? -->\n\n## Target Architecture\n\n\u003C!-- What does the new architecture look like? Key decisions and rationale. -->\n\n### Key Decisions\n\n\u003C!-- Tag each as KD-N for cross-referencing -->\n```\n\n### 5-interfaces-apis.md\n\n```markdown\n# §5 — Interfaces and APIs\n\n> External APIs, internal endpoints, database schemas.\n\n## External APIs\n\n\u003C!-- APIs consumed from third parties -->\n\n## Internal APIs\n\n\u003C!-- Endpoints this system exposes -->\n\n## Database Schema\n\n\u003C!-- Key tables\u002Fcollections and their structure -->\n```\n\n### 6-implementation-plan.md\n\n```markdown\n# §6 — Implementation Plan\n\n> Phases, workstreams, prompts, tests, migrations, rollback.\n\n## 6.1 Phased Rollout\n\n\u003C!-- Build order, dependencies, what's parallelizable -->\n\n## 6.2 Workstream Breakdown\n\n\u003C!-- Parallel work tracks, interface contracts between them -->\n\n## 6.3 Prompt Sequence\n\n\u003C!-- One prompt per implementation unit -->\n\u003C!--\nEach prompt should include:\n- \u003Cread-first> block listing 4-6 specific source files\n- The existing \u002Fplan workflow; recommend an explicit reasoning effort only when\n  the active runtime supports that control. Do not invent a slash command.\n- TDD test names (the test is the first deliverable)\n- Exact file paths to create\u002Fmodify\n- A \u003Cproof> block: the tests, the command that runs the new behavior, what the\n  run must show, and the done_when list (references\u002Fcc-prompt-templates.md).\n  Each unit is one \u002Fdeliver contract.\n-->\n\n## 6.4 Testing Strategy\n\n\u003C!--\nFor each kind below: name where it applies, or write \"N\u002FA — \u003Creason>\".\nAn unexplained blank is not an answer.\n\n  · unit            — per-function behavior\n  · contract        — payload\u002FAPI invariants that must never silently break\n  · property        — invariants across the input space, not per-case\n  · golden\u002Ffixture  — locks behavior against a validated reference\n  · integration     — real dependencies; replay against historical data\n  · mutation        — proves the suite constrains the code; use wherever the\n                      code and its tests share an author, human or agent\n  · statistical     — REQUIRED when the output is probabilistic, ranked, or\n                      scored: name the metric, the baseline, and the sample size\n  · load\u002Ftorture    — behavior at full scale\n  · post-deploy     — what reality reports back after shipping\n  · QA (run it)     — REQUIRED per unit: the command, page, or call that runs\n                      the new behavior, and what it must show; \"N\u002FA\" only when\n                      the unit has no runnable surface\n-->\n\n| Unit | Kinds applied | QA (run it) | N\u002FA + reason |\n| ---- | ------------- | ----------- | ------------ |\n|      |               |             |              |\n\n## 6.5 Migration Sequence\n\n\u003C!-- Numbered steps, all additive (no breaking changes), deployable before code -->\n\n## 6.6 Rollback Plan\n\n\u003C!-- Scenario → action table, coexistence strategy for old\u002Fnew -->\n\n| Scenario | Action | Notes |\n| -------- | ------ | ----- |\n|          |        |       |\n```\n\n### 7-non-functional-requirements.md\n\n```markdown\n# §7 — Non-Functional Requirements\n\n> Performance, reliability, operational, security constraints.\n\n\u003C!-- A threshold constraint names its metric, its value, and what happens on breach.\n     \"Fast\", \"reliable\", \"a drift alarm\" are adjectives, not constraints.\n     An invariant needs no number — \"all writes are atomic\", \"never overwrites a\n     user file\" are binary and testable exactly as written. Reach for a number\n     only where you are promising a threshold.\n     A measurement recorded in research\u002F is not a constraint until it lands here.\n     `dotbabel-validate-specs` enforces this: a comparative with no value fails. -->\n\n## Performance\n\n\u003C!-- Tag each as PERF-N -->\n\n## Reliability\n\n\u003C!-- Tag each as REL-N -->\n\n## Operational\n\n\u003C!-- Tag each as OPS-N -->\n\n## Security\n\n\u003C!-- Tag each as SEC-N -->\n```\n\n### 8-risks-alternatives.md\n\n```markdown\n# §8 — Risks and Alternatives\n\n> Known risks with mitigations, rejected approaches with reasoning.\n\n## Risks\n\n\u003C!-- Tag each as R-N -->\n\n| ID  | Risk | Likelihood | Impact | Mitigation |\n| --- | ---- | ---------- | ------ | ---------- |\n|     |      |            |        |            |\n\n## Rejected Alternatives\n\n\u003C!-- Tag each as A-N. Include WHY it was rejected — future readers need the reasoning. -->\n```\n\n### spec.json\n\nThe machine-readable half of the spec. `dotbabel-validate-specs` requires\n`id`, `title`, `status`, `owners`, `linked_paths` and `acceptance_commands`;\n`acceptance_criteria` is optional but is what makes the spec's claims\nmachine-checkable rather than prose.\n\n```json\n{\n  \"id\": \"\u003Cspec-name>\",\n  \"title\": \"\u003CSpec Title>\",\n  \"status\": \"draft\",\n  \"owners\": [\"@\u003Cowner>\"],\n  \"linked_paths\": [\"\u003Cpath\u002Fthis\u002Fspec\u002Fgoverns>\"],\n  \"acceptance_commands\": [\"\u003Cthe command that proves this spec works>\"],\n  \"depends_on_specs\": [],\n  \"active_prs\": [],\n  \"acceptance_criteria\": [\n    {\n      \"id\": \"AC-1\",\n      \"status\": \"planned\",\n      \"given\": \"\u003Cthe precondition, in the system's own terms>\",\n      \"when\": \"\u003Cthe action taken>\",\n      \"then\": \"\u003Cthe observable outcome — one claim, not a list>\",\n      \"tests\": [\n        {\n          \"file\": \"\u003Cpath\u002Fto\u002Fthe.test.file>\",\n          \"name\": \"\u003Cthe exact test name, copied verbatim from the test>\"\n        }\n      ],\n      \"argv\": [\n        \"\u003Cthe command that runs this test>\",\n        \"\u003Cits arguments>\",\n        \"\u003Cpath\u002Fto\u002Fthe.test.file>\",\n        \"--reporter=junit\",\n        \"--outputFile=.dotbabel\u002Fcriteria\u002FAC-1.junit.xml\"\n      ],\n      \"report\": { \"format\": \"junit-xml\", \"path\": \".dotbabel\u002Fcriteria\u002FAC-1.junit.xml\" }\n    }\n  ]\n}\n```\n\n**Every new criterion starts `planned`.** A criterion becomes `active` only in\nthe pull request that adds the tests it names — seeding `active` at scaffold\ntime asserts, from the first commit, that tests exist which do not, and the\nverification command will fail looking for them.\n\n`linked_paths` and `acceptance_commands` must both be **non-empty** — the\nvalidator rejects an empty array for either, so replace the placeholders rather\nthan deleting them. `depends_on_specs` and `active_prs` must be present but may\nstay empty.\n\nKeep `linked_paths` narrow. A broad entry (`package.json`, a whole `src\u002F`)\ndrags unrelated changes into this spec's merge gate, so add a path only when a\nchange there genuinely needs this spec's criteria to pass.\n\nLeave `tests[].name` and `argv` as placeholders until the tests exist. The name\nmust later match the test **verbatim** — verification looks it up by exact\nstring in the JUnit report, so a paraphrase silently never matches.\n\n`argv` is runner-agnostic: the only fixed requirement is that the command emits\nJUnit XML at `report.path`, which every major runner can do. For a Node project\nthat is `[\"npx\", \"vitest\", \"run\", \"\u003Cfile>\", \"--reporter=junit\",\n\"--outputFile=…\"]`; Go uses `gotestsum --junitfile`, Python `pytest\n--junitxml`. Write the one your project actually runs rather than copying the\nNode form.\n\n### README.md\n\n```markdown\n# {Spec Name} — Engineering Spec\n\n> {description from $1, or omit this line if none provided}\n>\n> Created: {date}\n\n## Status\n\n| #   | Section                     | Status    |\n| --- | --------------------------- | --------- |\n| 1   | Problem \u002F Motivation        | [ ] empty |\n| 2   | Scope                       | [ ] empty |\n| 3   | High-Level Architecture     | [ ] empty |\n| 4   | Data Flow \u002F Components      | [ ] empty |\n| 5   | Interfaces and APIs         | [ ] empty |\n| 6   | Implementation Plan         | [ ] empty |\n| 7   | Non-Functional Requirements | [ ] empty |\n| 8   | Risks and Alternatives      | [ ] empty |\n\n## Quick Start\n\n\u003C!-- Filled in during finalization -->\n\n## Research Sources\n\nSee [research\u002Fsources.md](research\u002Fsources.md) for indexed source documents.\n```\n\n### research\u002Fsources.md\n\n```markdown\n# Research Sources\n\n> Indexed documents feeding into this spec. Each tagged with which sections it informs.\n\n\u003C!-- Format:\n- **DOC-N**: {title} — {one-line description}. Feeds: §N, §N.\n-->\n```\n\n### current-state\u002Fanalysis.md (brownfield only)\n\n```markdown\n# Current State Analysis\n\n> Analysis of the existing system being redesigned.\n\n## System Overview\n\n\u003C!-- What exists today? -->\n\n## Pain Points\n\n\u003C!-- What's broken, slow, or hard to maintain? -->\n\n## Preserved Behaviors\n\n\u003C!-- What MUST remain the same after the rewrite? -->\n```\n\n---\n\n## Constraint ID Conventions\n\n| Prefix | Domain         | Lives In            |\n| ------ | -------------- | ------------------- |\n| ARCH-N | Architecture   | §3, §4              |\n| IMPL-N | Implementation | §6                  |\n| TEST-N | Testing        | §6                  |\n| PERF-N | Performance    | §7                  |\n| REL-N  | Reliability    | §7                  |\n| OPS-N  | Operational    | §7                  |\n| SEC-N  | Security       | §7                  |\n| KD-N   | Key Decisions  | §4 target arch      |\n| R-N    | Risks          | §8                  |\n| A-N    | Alternatives   | §8                  |\n| DOC-N  | Research docs  | research\u002Fsources.md |\n\n---\n\n## Key Principles\n\n1. **User drives content, Claude drives structure.** Never invent requirements. Place user input in the correct section with proper tagging.\n2. **Scaffold first, fill interactively.** Create the empty structure immediately so the user can see what they're building toward.\n3. **Cross-reference everything.** Constraints reference sections. Docs reference sections. Prompts reference docs and files.\n4. **TDD-first in §6.** Every prompt in the Implementation Plan must list failing test names before implementation instructions.\n5. **Empty sections are fine.** Don't fill sections preemptively. Mark them empty and move on.\n6. **Constraints belong where they shape design.** A testing requirement goes in §6, not §7. A modularity requirement goes in §4, not §7. Operational limits go in §7.\n7. **One question at a time.** Don't overwhelm the user with multiple questions. Ask about one section or one aspect, then write what they give you.\n",{"name":5,"description":48,"argument-hint":49},"Create structured engineering specs through interactive pairing. Use when the user wants to create a spec, design doc, technical specification, architecture document, RFC, or engineering plan. Triggers on \"let's spec this out\", \"create a spec\", \"design doc\", \"write a technical plan\", \"plan the architecture\". Works for greenfield and brownfield projects. Outputs to docs\u002Fspecs\u002F.\n","[spec-name] [description] [--brownfield]","main","91c337164997b99943fbcc64e8e362577b67bc81","plugins\u002Fdotbabel\u002Ftemplates\u002Fclaude\u002Fskills\u002Fspec","https:\u002F\u002Fgithub.com\u002Fkaiohenricunha\u002Fdotbabel","https:\u002F\u002Fgithub.com\u002Fkaiohenricunha","I'm a Platform\u002FInfra engineer who happens to work in cybersec and write Go code.","2026-10-26 14:12:51",{"source":8,"skillName":5,"commands":58,"prompt":68,"zipAvailable":69},[59,62,65],{"tool":60,"command":61},"npx","npx skills add kaiohenricunha\u002Fdotbabel --skill spec",{"tool":63,"command":64},"bunx","bunx skills add kaiohenricunha\u002Fdotbabel --skill spec",{"tool":66,"command":67},"pnpm","pnpm dlx skills add kaiohenricunha\u002Fdotbabel --skill spec","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fkaiohenricunha\u002Fdotbabel\u002Fblob\u002Fmain\u002Fplugins\u002Fdotbabel\u002Ftemplates\u002Fclaude\u002Fskills\u002Fspec\u002FSKILL.md",true,[71],{"path":72,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":75,"sourceHtmlUrl":76},"references\u002Fcc-prompt-templates.md","REFERENCE","参考文档",3932,"https:\u002F\u002Fgithub.com\u002Fkaiohenricunha\u002Fdotbabel\u002Fblob\u002Fmain\u002Fplugins\u002Fdotbabel\u002Ftemplates\u002Fclaude\u002Fskills\u002Fspec\u002Freferences\u002Fcc-prompt-templates.md",[78,108,139,160,182,201],{"skillNo":79,"name":80,"nameZh":6,"description":81,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":82,"categories":83,"keywords":89,"publishedAt":43},"01M6XDQ7QMASPFZ355MYKCD06T","yolo","Start temporary autonomy for the current task or an explicit scope with a required --for expiry. Codex permits scoped pushes, merges, deploys, and releases after strict testing and validation; Full access is a separate runtime setting. Claude Code uses a session permission hook with a protected-action denylist.","https:\u002F\u002Fgithub.com\u002Fkaiohenricunha\u002Fdotbabel\u002Fblob\u002Fmain\u002Fskills\u002Fyolo\u002FSKILL.md",[84],{"categoryNo":85,"categoryName":86,"taxonomyType":87,"taxonomyTypeDesc":88},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[80,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,28,107],"testing","merge","code","start","access","session","ops","autonomy","releases","permission","full","runtime","scope","devex","hook","true","action","scoped",{"skillNo":109,"name":110,"nameZh":6,"description":111,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":112,"categories":113,"keywords":120,"publishedAt":43},"01M6XDQ1N21VE4KH98JWGZ7GN9","veracity-audit","Audit a data pipeline for Veracity and Value. Dispatches data-scientist, compliance-auditor, and data-engineer agents with project context injected at dispatch time. All source paths are supplied via flags — no defaults, no project assumptions. Subcommands: audit (full pipeline walk), score-check (scoring math only), gate-check (gate coverage only), source-trace \u003Clabel> (single source end-to-end). Invoke when: \"audit pipeline\", \"veracity check\", \"scoring math correct?\", \"check gates\", \"trace source\", \"verify quality gates\", \"formula correct?\".","https:\u002F\u002Fgithub.com\u002Fkaiohenricunha\u002Fdotbabel\u002Fblob\u002Fmain\u002Fskills\u002Fveracity-audit\u002FSKILL.md",[114,117],{"categoryNo":115,"categoryName":116,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",{"categoryNo":118,"categoryName":119,"taxonomyType":87,"taxonomyTypeDesc":88},"01JBSEEDCAT000000000000037","律师与法务",[121,122,90,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,24],"audit","veracity audit","review","data pipeline","pipeline","Bash","math","coverage","only","engineer","formula","scoring","time","project","worker","end","at","invoke",{"skillNo":140,"name":141,"nameZh":6,"description":142,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":143,"categories":144,"keywords":145,"publishedAt":43},"01M6XDPVDKBR376C06CEK9BC5S","validate-spec","Audit an already-implemented spec against the codebase. Walks each constraint (ARCH-N, PERF-N, KD-N, etc.) and acceptance criterion, grounds findings in file:line evidence, runs the spec.json acceptance_commands, and writes a single audit doc to docs\u002Faudits\u002F. Use whenever the user asks to \"validate a spec\", \"audit a spec\", \"check if spec is implemented\", \"verify the spec is done\", \"is this spec really done\", or otherwise wants closure on spec-driven work. Read-only against the spec — produces an audit, never modifies the spec itself.","https:\u002F\u002Fgithub.com\u002Fkaiohenricunha\u002Fdotbabel\u002Fblob\u002Fmain\u002Fskills\u002Fvalidate-spec\u002FSKILL.md",[],[5,146,147,90,123,148,149,121,126,150,151,152,129,153,154,155,156,157,158,159],"validate","validate spec","run","acceptance","work","read","never","max","done","arch","line","2025","spec-driven","closure",{"skillNo":161,"name":162,"nameZh":6,"description":163,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":164,"categories":165,"keywords":169,"publishedAt":43},"01M6XDPMMECH82RAW553Y94HDS","terragrunt-specialist","Deep-dive Terragrunt hierarchy review, DRY pattern audits, and run-all orchestration analysis. Use for structured investigations of multi-environment Terragrunt layouts, dependency graphs, remote state config, and hook correctness. Triggers on: \"Terragrunt audit\", \"run-all review\", \"dependency block\", \"DRY pattern review\", \"env hierarchy audit\", \"mock_outputs\", \"terragrunt hooks\".","https:\u002F\u002Fgithub.com\u002Fkaiohenricunha\u002Fdotbabel\u002Fblob\u002Fmain\u002Fskills\u002Fterragrunt-specialist\u002FSKILL.md",[166],{"categoryNo":167,"categoryName":168,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000011","运维与部署",[170,171,172,123,173,174,148,121,175,126,151,176,177,153,178,179,157,180,137,181],"terragrunt","specialist","terragrunt specialist","orchestration","debugging","analysis","dependency","hooks","deep","dry","remote","hierarchy",{"skillNo":183,"name":184,"nameZh":6,"description":185,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":186,"categories":187,"keywords":193,"publishedAt":43},"01M6XDPFJS2HGSE13RXF1HV8S7","terraform-specialist","Deep-dive Terraform architecture review, module design, state management, and migration. Use for structured investigations of Terraform workspaces, provider configuration, module coupling, import workflows, and test coverage. Triggers on: \"Terraform audit\", \"module review\", \"state management\", \"Terraform import\", \"workspace design\", \"provider config review\", \"Terraform testing\".","https:\u002F\u002Fgithub.com\u002Fkaiohenricunha\u002Fdotbabel\u002Fblob\u002Fmain\u002Fskills\u002Fterraform-specialist\u002FSKILL.md",[188,191,192],{"categoryNo":189,"categoryName":190,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000008","测试与质量",{"categoryNo":167,"categoryName":168,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[194,195,171,90,25,123,196,42,197,174,121,126,151,128,198,199,178,200,157,137],"terraform","terraform specialist","workflows","architecture review","test coverage","workspaces","problem",{"skillNo":202,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":203,"categories":204,"keywords":209,"publishedAt":43},"01M6XDP7V6H8WYPSXVEKYE6HJ6","https:\u002F\u002Fgithub.com\u002Fkaiohenricunha\u002Fdotbabel\u002Fblob\u002Fmain\u002Fskills\u002Fspec\u002FSKILL.md",[205,206],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":207,"categoryName":208,"taxonomyType":87,"taxonomyTypeDesc":88},"01JBSEEDCAT000000000000032","设计师",[5,25,210,42,41,39,38,35,211,34,33,32,30,212,29,28,213,214,26,215],"component","interactive","opus","trade-offs","technical plan","frontier"]