[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6XDGYKMQVH5A0ZWP86PST25":3,"skill-same:01M6XDGYKMQVH5A0ZWP86PST25":84},{"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":18,"publishedAt":39,"linkOnly":40,"markdownContent":41,"markdownBody":42,"frontMatter":43,"refName":63,"commitSha":64,"skillDirectory":65,"forksCount":14,"repositoryHtmlUrl":66,"repositoryHomepageUrl":6,"ownerHtmlUrl":67,"ownerBlogUrl":6,"ownerDescription":68,"updatedAt":69,"install":70,"resources":83},"01M6XDGYKMQVH5A0ZWP86PST25","create-experiment",null,"Run a scoped, local-only experiment and save the report to docs\u002Fexperiments\u002F. Triggers on: \"let's try\", \"compare A vs B\", \"prototype\", \"explore\".","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\u002Fskills\u002Fcreate-experiment\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38],"create experiment","create","experiment","write","routine","scoped","sonnet","devex","local","wed","only","review","read","explore","Bash","save","run","documentation","experiments","workflow","2026-11-10 10:12:30",false,"---\nid: create-experiment\nname: create-experiment\ntype: skill\nversion: 1.0.0\ndomain:\n  - devex\nplatform:\n  - none\ntask:\n  - review\n  - documentation\nmaturity: validated\nowner: \"@kaiohenricunha\"\ncreated: 2026-04-29T00:00:00.000Z\nupdated: 2026-04-30T00:00:00.000Z\ndescription: |\n  Run a scoped, local-only experiment and save the report to docs\u002Fexperiments\u002F. Triggers on: \"let's try\", \"compare A vs B\", \"prototype\", \"explore\".\nargument-hint: \"[topic or hypothesis]\"\ntools: Read, Grep, Glob, Bash, Write\nallowed-tools: Read Grep Glob Bash Write\nmodel: sonnet\ndotbabel:\n  compute:\n    mode: dynamic\n    requirement: routine\n    binding: consumer\n    rationale: Scoped engineering follows an existing implementation or review workflow with checkable results.\n---\n\nRun a scoped, local-only experiment and produce a structured report saved to the project's `docs\u002Fexperiments\u002F` directory.\n\nTrigger: when the user asks \"let's try X\", \"I want to compare A vs B\", \"can we prototype Y\", \"explore Z before we commit\", or invokes `\u002Fcreate-experiment` directly. Use this skill **before** `\u002Fspec` (which is for design docs, not exploratory probes) and **after** plain shell tinkering becomes too messy to track.\n\nArguments: `$ARGUMENTS` — a topic or hypothesis (e.g. \"ripgrep vs grep on this repo\", \"switch from pnpm to bun\", \"does Postgres LISTEN\u002FNOTIFY scale to N clients\"). Required — if empty, ask the user what they want to explore.\n\n## Purpose\n\n`\u002Fcreate-experiment` is **not** a spec (no formal design), **not** an audit (it doesn't enumerate issues), **not** a fix (it doesn't ship anything). It is a **decision-grade probe**: refine a hypothesis, run it locally, capture results — including negative ones — and recommend a next move.\n\nThe output is a dated markdown file under `docs\u002Fexperiments\u002F` plus a sandbox directory containing the runnable artifacts. Both are left untracked for the user to review.\n\n## Steps\n\n### Step 0 — Refine the goal (interactive, blocking)\n\nIf `$ARGUMENTS` already covers hypothesis + observable success signal, echo the refined goal back as a paragraph + bulleted success criteria and **wait for user sign-off** before proceeding. Skip questions already answered in `$ARGUMENTS` or recent context.\n\nOtherwise, ask up to 4 short questions:\n\n1. **What are you trying to learn?** — one-sentence hypothesis.\n2. **What does \"it worked\" look like?** — concrete, observable signal (a benchmark number, a working prototype, a config that boots, a passing test).\n3. **What's out of scope?** — what should NOT be touched.\n4. **Time-box** — sketch (≤30 min), half-day (≤4 hr), or full-day (≤8 hr).\n\nEcho the refined goal back as a paragraph + bulleted success criteria. **Wait for explicit user sign-off before continuing.** Do not proceed to Step 1 until the user confirms.\n\n### Step 1 — Plan the experiment\n\n- Propose 1–3 approaches to try. Each is a distinct path to the same hypothesis.\n- For each approach, sketch: what gets installed, what runs, what's measured.\n- Pick the sandbox target:\n  - **Default:** a fresh git worktree at `.claude\u002Fworktrees\u002Fexperiment-\u003Cslug>\u002F` branched from the latest `origin\u002Fmain`. Run `git fetch origin main` first.\n  - **Fallback:** `~\u002Fexperiments\u002F\u003Cslug>\u002F` if there is no enclosing git repo.\n- Show this plan to the user and get a final go-ahead before touching the system.\n\n### Step 2 — Set up the environment\n\nExecute setup commands in the sandbox. **Capture every command and its output** as the reproducibility ledger — redact any tokens or credentials before including output in the report. Examples:\n\n- `git worktree add .claude\u002Fworktrees\u002Fexperiment-\u003Cslug> -b experiment\u002F\u003Cslug> origin\u002Fmain`\n- `npm i \u003Cpkg>` \u002F `pnpm add \u003Cpkg>` \u002F `bun add \u003Cpkg>`\n- `python -m venv .venv && source .venv\u002Fbin\u002Factivate && pip install -r requirements.txt`\n- `docker compose up -d`\n- `cargo new \u003Cname> && cd \u003Cname>`\n\nIf a setup step fails, document the failure, attempt **one** recovery (e.g. install a missing system dep), and proceed only if it succeeds. Do not silently swallow errors.\n\n### Step 3 — Execute the approaches\n\nApproaches are independent by default — execute concurrently when possible (parallel tool calls or background jobs with per-approach log files). Fall back to serial execution only when one approach depends on another's artifact.\n\nFor each approach:\n\n- Save code\u002Fconfig under `experiments\u002F\u003Cslug>\u002F\u003Capproach>\u002F` inside the sandbox.\n- Run the approach. For measurement commands and any failing command, capture full output — at minimum the last 10 lines.\n- Evaluate against the success criteria from Step 0. Record the result as **PASS**, **PARTIAL**, or **FAIL**.\n- If an approach surfaces a blocker that invalidates the hypothesis itself, stop and surface that to the user before continuing — they may want to refine the goal.\n\n### Step 4 — Compare & recommend\n\nBuild a comparison table that maps each approach against the agreed success criteria. Pick a recommendation, **or** explicitly say \"none of these — here's why and what to try next.\" A clean negative result is a complete experiment.\n\n### Step 5 — Write the report\n\nGenerate a filename: `\u003Ctopic-slug>-\u003CYYYY-MM-DD>.md` in lowercase kebab-case (e.g. `ripgrep-vs-grep-2026-04-29.md`). Create `docs\u002Fexperiments\u002F` if it doesn't exist. Use this structure:\n\n````markdown\n# Experiment: \u003CTopic> — \u003CYYYY-MM-DD>\n\n\u003COne-sentence hypothesis.>\n\n## Goal\n\n\u003C2–3 sentences. What we're learning and why this experiment was run now.>\n\n## Success Criteria\n\n- \u003Cobservable signal 1>\n- \u003Cobservable signal 2>\n\n## Environment Setup\n\nSandbox: `\u003Cabsolute path chosen in Step 1>`\n\n```bash\n\u003Cexact commands run, in order>\n```\n\n\u003Coutput snippets for any measurement or failing command — last 10 lines minimum>\n\n## Approaches Tried\n\n### Approach 1: \u003CName>\n\n- **Idea:** \u003Cone line>\n- **Code\u002Fconfig:** `\u003Cpath-in-sandbox>`\n- **Commands:**\n\n  ```bash\n  \u003Ccommands>\n  ```\n\n- **Result:** PASS | PARTIAL | FAIL — \u003Cobserved signal vs criterion>\n- **Notes:** \u003Cgotchas, surprises, dead-ends>\n\n\u003C!-- Add sections for each approach tried; omit if only one approach was attempted -->\n\n## Comparison\n\n| Approach | Result | \u003CCriterion 1> | \u003CCriterion 2> | Effort |\n| -------- | ------ | ------------- | ------------- | ------ |\n| ...      | ...    | ...           | ...           | ...    |\n\n## Recommendation\n\n**\u003CAdopt Approach N | Iterate further | Abandon>**\n\n\u003C2–3 sentences. Why this conclusion best matches the success criteria. Call out\nassumptions or unresolved questions.>\n\n## Next Step\n\n- Promote to `\u002Fspec \u003Ctopic>` to formalize the chosen approach, OR\n- Run `\u002Ffix-with-evidence \u003Ctopic>` to implement directly, OR\n- Run `\u002Freproduce-bug \u003Csymptom>` to capture a surprising failure as a failing regression test, OR\n- Re-run `\u002Fcreate-experiment` with a refined hypothesis, OR\n- Drop it — this document is the negative-result record.\n\n## Sandbox Cleanup\n\n```bash\ngit worktree remove .claude\u002Fworktrees\u002Fexperiment-\u003Cslug>   # or\nrm -rf ~\u002Fexperiments\u002F\u003Cslug>\n```\n````\n\n### Step 6 — Report back\n\nReply with: doc path, one-sentence recommendation, and sandbox path. **Do not paste the full document into chat.**\n\n## Rules\n\n- All work runs **locally**. No production endpoints, no cloud writes, no real auth tokens, no shared infrastructure. If the experiment needs an external service, mock it or use a disposable test account.\n- Environment setup is part of the experiment — every install\u002Fconfig\u002Fboot command must appear in the report. Reproducibility is the bar.\n- Capture failed approaches too. Negative results are valuable signal and **must** be in the report.\n- Do not commit the experiment doc or the sandbox. Leave both untracked for the user.\n- Do not install global tools (`apt`, `brew`, `npm i -g`) without explicit user approval during Step 1.\n- Tables and code blocks over prose. No filler.\n- An experiment is **done when success criteria are evaluated** — not when \"everything works.\" A clean failure is a finished experiment.\n","Run a scoped, local-only experiment and produce a structured report saved to the project's `docs\u002Fexperiments\u002F` directory.\n\nTrigger: when the user asks \"let's try X\", \"I want to compare A vs B\", \"can we prototype Y\", \"explore Z before we commit\", or invokes `\u002Fcreate-experiment` directly. Use this skill **before** `\u002Fspec` (which is for design docs, not exploratory probes) and **after** plain shell tinkering becomes too messy to track.\n\nArguments: `$ARGUMENTS` — a topic or hypothesis (e.g. \"ripgrep vs grep on this repo\", \"switch from pnpm to bun\", \"does Postgres LISTEN\u002FNOTIFY scale to N clients\"). Required — if empty, ask the user what they want to explore.\n\n## Purpose\n\n`\u002Fcreate-experiment` is **not** a spec (no formal design), **not** an audit (it doesn't enumerate issues), **not** a fix (it doesn't ship anything). It is a **decision-grade probe**: refine a hypothesis, run it locally, capture results — including negative ones — and recommend a next move.\n\nThe output is a dated markdown file under `docs\u002Fexperiments\u002F` plus a sandbox directory containing the runnable artifacts. Both are left untracked for the user to review.\n\n## Steps\n\n### Step 0 — Refine the goal (interactive, blocking)\n\nIf `$ARGUMENTS` already covers hypothesis + observable success signal, echo the refined goal back as a paragraph + bulleted success criteria and **wait for user sign-off** before proceeding. Skip questions already answered in `$ARGUMENTS` or recent context.\n\nOtherwise, ask up to 4 short questions:\n\n1. **What are you trying to learn?** — one-sentence hypothesis.\n2. **What does \"it worked\" look like?** — concrete, observable signal (a benchmark number, a working prototype, a config that boots, a passing test).\n3. **What's out of scope?** — what should NOT be touched.\n4. **Time-box** — sketch (≤30 min), half-day (≤4 hr), or full-day (≤8 hr).\n\nEcho the refined goal back as a paragraph + bulleted success criteria. **Wait for explicit user sign-off before continuing.** Do not proceed to Step 1 until the user confirms.\n\n### Step 1 — Plan the experiment\n\n- Propose 1–3 approaches to try. Each is a distinct path to the same hypothesis.\n- For each approach, sketch: what gets installed, what runs, what's measured.\n- Pick the sandbox target:\n  - **Default:** a fresh git worktree at `.claude\u002Fworktrees\u002Fexperiment-\u003Cslug>\u002F` branched from the latest `origin\u002Fmain`. Run `git fetch origin main` first.\n  - **Fallback:** `~\u002Fexperiments\u002F\u003Cslug>\u002F` if there is no enclosing git repo.\n- Show this plan to the user and get a final go-ahead before touching the system.\n\n### Step 2 — Set up the environment\n\nExecute setup commands in the sandbox. **Capture every command and its output** as the reproducibility ledger — redact any tokens or credentials before including output in the report. Examples:\n\n- `git worktree add .claude\u002Fworktrees\u002Fexperiment-\u003Cslug> -b experiment\u002F\u003Cslug> origin\u002Fmain`\n- `npm i \u003Cpkg>` \u002F `pnpm add \u003Cpkg>` \u002F `bun add \u003Cpkg>`\n- `python -m venv .venv && source .venv\u002Fbin\u002Factivate && pip install -r requirements.txt`\n- `docker compose up -d`\n- `cargo new \u003Cname> && cd \u003Cname>`\n\nIf a setup step fails, document the failure, attempt **one** recovery (e.g. install a missing system dep), and proceed only if it succeeds. Do not silently swallow errors.\n\n### Step 3 — Execute the approaches\n\nApproaches are independent by default — execute concurrently when possible (parallel tool calls or background jobs with per-approach log files). Fall back to serial execution only when one approach depends on another's artifact.\n\nFor each approach:\n\n- Save code\u002Fconfig under `experiments\u002F\u003Cslug>\u002F\u003Capproach>\u002F` inside the sandbox.\n- Run the approach. For measurement commands and any failing command, capture full output — at minimum the last 10 lines.\n- Evaluate against the success criteria from Step 0. Record the result as **PASS**, **PARTIAL**, or **FAIL**.\n- If an approach surfaces a blocker that invalidates the hypothesis itself, stop and surface that to the user before continuing — they may want to refine the goal.\n\n### Step 4 — Compare & recommend\n\nBuild a comparison table that maps each approach against the agreed success criteria. Pick a recommendation, **or** explicitly say \"none of these — here's why and what to try next.\" A clean negative result is a complete experiment.\n\n### Step 5 — Write the report\n\nGenerate a filename: `\u003Ctopic-slug>-\u003CYYYY-MM-DD>.md` in lowercase kebab-case (e.g. `ripgrep-vs-grep-2026-04-29.md`). Create `docs\u002Fexperiments\u002F` if it doesn't exist. Use this structure:\n\n````markdown\n# Experiment: \u003CTopic> — \u003CYYYY-MM-DD>\n\n\u003COne-sentence hypothesis.>\n\n## Goal\n\n\u003C2–3 sentences. What we're learning and why this experiment was run now.>\n\n## Success Criteria\n\n- \u003Cobservable signal 1>\n- \u003Cobservable signal 2>\n\n## Environment Setup\n\nSandbox: `\u003Cabsolute path chosen in Step 1>`\n\n```bash\n\u003Cexact commands run, in order>\n```\n\n\u003Coutput snippets for any measurement or failing command — last 10 lines minimum>\n\n## Approaches Tried\n\n### Approach 1: \u003CName>\n\n- **Idea:** \u003Cone line>\n- **Code\u002Fconfig:** `\u003Cpath-in-sandbox>`\n- **Commands:**\n\n  ```bash\n  \u003Ccommands>\n  ```\n\n- **Result:** PASS | PARTIAL | FAIL — \u003Cobserved signal vs criterion>\n- **Notes:** \u003Cgotchas, surprises, dead-ends>\n\n\u003C!-- Add sections for each approach tried; omit if only one approach was attempted -->\n\n## Comparison\n\n| Approach | Result | \u003CCriterion 1> | \u003CCriterion 2> | Effort |\n| -------- | ------ | ------------- | ------------- | ------ |\n| ...      | ...    | ...           | ...           | ...    |\n\n## Recommendation\n\n**\u003CAdopt Approach N | Iterate further | Abandon>**\n\n\u003C2–3 sentences. Why this conclusion best matches the success criteria. Call out\nassumptions or unresolved questions.>\n\n## Next Step\n\n- Promote to `\u002Fspec \u003Ctopic>` to formalize the chosen approach, OR\n- Run `\u002Ffix-with-evidence \u003Ctopic>` to implement directly, OR\n- Run `\u002Freproduce-bug \u003Csymptom>` to capture a surprising failure as a failing regression test, OR\n- Re-run `\u002Fcreate-experiment` with a refined hypothesis, OR\n- Drop it — this document is the negative-result record.\n\n## Sandbox Cleanup\n\n```bash\ngit worktree remove .claude\u002Fworktrees\u002Fexperiment-\u003Cslug>   # or\nrm -rf ~\u002Fexperiments\u002F\u003Cslug>\n```\n````\n\n### Step 6 — Report back\n\nReply with: doc path, one-sentence recommendation, and sandbox path. **Do not paste the full document into chat.**\n\n## Rules\n\n- All work runs **locally**. No production endpoints, no cloud writes, no real auth tokens, no shared infrastructure. If the experiment needs an external service, mock it or use a disposable test account.\n- Environment setup is part of the experiment — every install\u002Fconfig\u002Fboot command must appear in the report. Reproducibility is the bar.\n- Capture failed approaches too. Negative results are valuable signal and **must** be in the report.\n- Do not commit the experiment doc or the sandbox. Leave both untracked for the user.\n- Do not install global tools (`apt`, `brew`, `npm i -g`) without explicit user approval during Step 1.\n- Tables and code blocks over prose. No filler.\n- An experiment is **done when success criteria are evaluated** — not when \"everything works.\" A clean failure is a finished experiment.\n",{"id":5,"name":5,"task":44,"type":45,"model":25,"owner":46,"tools":47,"domain":48,"created":49,"updated":50,"version":51,"dotbabel":52,"maturity":57,"platform":58,"description":60,"allowed-tools":61,"argument-hint":62},[30,36],"skill","@kaiohenricunha","Read, Grep, Glob, Bash, Write",[26],1777420800000,1777507200000,"1.0.0",{"compute":53},{"mode":54,"binding":55,"rationale":56,"requirement":23},"dynamic","consumer","Scoped engineering follows an existing implementation or review workflow with checkable results.","validated",[59],"none","Run a scoped, local-only experiment and save the report to docs\u002Fexperiments\u002F. Triggers on: \"let's try\", \"compare A vs B\", \"prototype\", \"explore\".\n","Read Grep Glob Bash Write","[topic or hypothesis]","main","91c337164997b99943fbcc64e8e362577b67bc81","skills\u002Fcreate-experiment","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":71,"prompt":81,"zipAvailable":82},[72,75,78],{"tool":73,"command":74},"npx","npx skills add kaiohenricunha\u002Fdotbabel --skill create-experiment",{"tool":76,"command":77},"bunx","bunx skills add kaiohenricunha\u002Fdotbabel --skill create-experiment",{"tool":79,"command":80},"pnpm","pnpm dlx skills add kaiohenricunha\u002Fdotbabel --skill create-experiment","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fkaiohenricunha\u002Fdotbabel\u002Fblob\u002Fmain\u002Fskills\u002Fcreate-experiment\u002FSKILL.md",true,[],[85,114,145,165,187,210],{"skillNo":86,"name":87,"nameZh":6,"description":88,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":89,"categories":90,"keywords":96,"publishedAt":39},"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",[91],{"categoryNo":92,"categoryName":93,"taxonomyType":94,"taxonomyTypeDesc":95},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[87,97,98,99,100,101,102,103,104,105,106,107,108,109,26,110,111,112,113,24],"testing","merge","code","start","access","session","ops","autonomy","releases","permission","full","runtime","scope","hook","true","action","for",{"skillNo":115,"name":116,"nameZh":6,"description":117,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":118,"categories":119,"keywords":128,"publishedAt":39},"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",[120,125],{"categoryNo":121,"categoryName":122,"taxonomyType":123,"taxonomyTypeDesc":124},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",{"categoryNo":126,"categoryName":127,"taxonomyType":94,"taxonomyTypeDesc":95},"01JBSEEDCAT000000000000037","律师与法务",[129,130,97,30,131,132,33,133,134,29,135,136,137,138,139,140,141,142,143,144],"audit","veracity audit","data pipeline","pipeline","math","coverage","engineer","formula","scoring","time","project","worker","end","at","invoke","when",{"skillNo":146,"name":147,"nameZh":6,"description":148,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":149,"categories":150,"keywords":151,"publishedAt":39},"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",[],[152,153,154,97,30,35,155,129,33,156,31,157,29,158,159,160,161,162,163,164],"spec","validate","validate spec","acceptance","work","never","max","done","arch","line","2025","spec-driven","closure",{"skillNo":166,"name":167,"nameZh":6,"description":168,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":169,"categories":170,"keywords":174,"publishedAt":39},"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",[171],{"categoryNo":172,"categoryName":173,"taxonomyType":123,"taxonomyTypeDesc":124},"01JBSEEDCAT000000000000011","运维与部署",[175,176,177,30,178,179,35,129,180,33,31,181,182,158,183,184,162,185,142,186],"terragrunt","specialist","terragrunt specialist","orchestration","debugging","analysis","dependency","hooks","deep","dry","remote","hierarchy",{"skillNo":188,"name":189,"nameZh":6,"description":190,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":191,"categories":192,"keywords":200,"publishedAt":39},"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",[193,196,197],{"categoryNo":194,"categoryName":195,"taxonomyType":123,"taxonomyTypeDesc":124},"01JBSEEDCAT000000000000008","测试与质量",{"categoryNo":172,"categoryName":173,"taxonomyType":123,"taxonomyTypeDesc":124},{"categoryNo":198,"categoryName":199,"taxonomyType":123,"taxonomyTypeDesc":124},"01JBSEEDCAT000000000000015","视觉设计",[201,202,176,97,203,30,204,205,206,179,129,33,31,134,207,208,183,209,162,142],"terraform","terraform specialist","design","workflows","architecture","architecture review","test coverage","workspaces","problem",{"skillNo":211,"name":152,"nameZh":6,"description":212,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":213,"categories":214,"keywords":219,"publishedAt":39},"01M6XDP7V6H8WYPSXVEKYE6HJ6","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.","https:\u002F\u002Fgithub.com\u002Fkaiohenricunha\u002Fdotbabel\u002Fblob\u002Fmain\u002Fskills\u002Fspec\u002FSKILL.md",[215,216],{"categoryNo":198,"categoryName":199,"taxonomyType":123,"taxonomyTypeDesc":124},{"categoryNo":217,"categoryName":218,"taxonomyType":94,"taxonomyTypeDesc":95},"01JBSEEDCAT000000000000032","设计师",[152,203,220,205,20,221,222,22,223,224,225,226,227,228,229,113,230,231,232,233],"component","plan","projects","interactive","use","document","specification","triggers","opus","technical specification","trade-offs","technical plan","outputs","frontier"]