[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6XDN1RGQVRCH8KQKBP5HQZB":3,"skill-same:01M6XDN1RGQVRCH8KQKBP5HQZB":87},{"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":44,"linkOnly":45,"markdownContent":46,"markdownBody":47,"frontMatter":48,"refName":66,"commitSha":67,"skillDirectory":68,"forksCount":14,"repositoryHtmlUrl":69,"repositoryHomepageUrl":6,"ownerHtmlUrl":70,"ownerBlogUrl":6,"ownerDescription":71,"updatedAt":72,"install":73,"resources":86},"01M6XDN1RGQVRCH8KQKBP5HQZB","reproduce-bug",null,"Reproduce a bug in an isolated, production-mirroring sandbox and capture a failing regression test as the objective baseline to fix against. Saves a report to docs\u002Freproductions\u002F. Triggers on: \"reproduce this bug\", \"can't reproduce\", \"make it fail reliably\", \"repro\".","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\u002Freproduce-bug\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000008","测试与质量","SCENARIO","使用场景",[24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43],"reproduce","bug","reproduce bug","failing","testing","sonnet","make","devex","as","isolation","triggers","deep","repro","fix","edit","write","read","regression","Bash","debugging","2026-11-10 10:12:30",false,"---\nid: reproduce-bug\nname: reproduce-bug\ntype: skill\nversion: 1.0.0\ndomain:\n  - devex\nplatform:\n  - none\ntask:\n  - debugging\n  - testing\nmaturity: validated\nowner: \"@kaiohenricunha\"\ncreated: 2026-06-14T00:00:00.000Z\nupdated: 2026-06-14T00:00:00.000Z\ndescription: |\n  Reproduce a bug in an isolated, production-mirroring sandbox and capture a failing regression test as the objective baseline to fix against. Saves a report to docs\u002Freproductions\u002F. Triggers on: \"reproduce this bug\", \"can't reproduce\", \"make it fail reliably\", \"repro\".\nargument-hint: \"[bug or symptom]\"\ntools: Read, Grep, Glob, Bash, Write, Edit\nmodel: sonnet\ndotbabel:\n  compute:\n    mode: dynamic\n    requirement: deep\n    binding: consumer\n    rationale: Isolation and regression capture must reproduce the actual triggering conditions.\n---\n\nReproduce a bug under isolation and the exact conditions that trigger it, then capture a **failing regression test** as the objective baseline to fix against. The output is a dated report under `docs\u002Freproductions\u002F` plus a failing test left in the sandbox.\n\nTrigger: when the user says \"reproduce this bug\", \"I can't reproduce X\", \"make it fail reliably\", \"repro Y\", or invokes `\u002Freproduce-bug` directly. Use it **after** a bug is reported (or surfaced by `\u002Fcreate-audit`) and **before** `\u002Ffix-with-evidence` writes the patch.\n\nArguments: `$ARGUMENTS` — a bug or symptom (e.g. \"checkout 500s when cart has >50 items\", \"intermittent panic in the scheduler\", issue link\u002Fnumber). If empty, ask the user what to reproduce.\n\n## Purpose\n\n`\u002Freproduce-bug` is **not** a fix (it ships no patch — that is `\u002Ffix-with-evidence`), **not** an audit (it doesn't enumerate issues — that is `\u002Fcreate-audit`), and **not** a flaky-test diagnosis of an existing suite (that is `\u002Fdetect-flaky`). It is the **bridge** between a bug report and a fix: gather the evidence, isolate a prod-mirroring environment, and produce a deterministically failing test.\n\nThe deliverable is a reproduction other skills build on. `\u002Ffix-with-evidence` consumes the failing test directly — it **satisfies that skill's Phase-1 Reproduce gate**, so the fix loop can start at Phase 2.\n\n## Steps\n\n### Step 0 — Gather the failure evidence (interactive, blocking)\n\nCollect the trigger conditions, echo them back, and **wait for sign-off** before touching anything. Skip what is already in `$ARGUMENTS` or recent context. If `$ARGUMENTS` is empty, ask what to reproduce.\n\n1. **Stack trace \u002F error** — the full message and the top frames. For a GitHub issue: `gh issue view \u003CN>`.\n2. **Inputs** — the exact user inputs, API payloads, CLI args, or dataset\u002Fuser-state that triggered the failure.\n3. **Environment** — OS, runtime\u002Flanguage version, framework, and DB\u002Fbrowser versions; the relevant env vars (redact secrets).\n4. **Logs & traces** — for web or distributed systems, the application logs \u002F telemetry that trace the execution path **before** the error was thrown.\n\nEcho back the consolidated evidence **and the precise failure condition you will target**. Wait for explicit user sign-off before continuing to Step 1.\n\n### Step 1 — Locate the code path\n\nUse `Grep` + `Read` to find the throwing path and the code that consumes the Step-0 inputs. Cite `file:line` for every claim. **Do not edit anything yet** — this is baseline-only; root cause is recorded later as candidates, not patched here.\n\n### Step 2 — Build the isolated, prod-mirroring sandbox\n\nIsolate the reproduction so nothing touches the main checkout, then mirror production along every axis the bug depends on.\n\n- **Sandbox:** `git fetch origin main` then `git worktree add .claude\u002Fworktrees\u002Frepro-\u003Cslug> -b repro\u002F\u003Cslug> origin\u002Fmain`. Fallback: `~\u002Frepro\u002F\u003Cslug>\u002F` if there is no enclosing git repo.\n- **Pin versions:** match the runtime\u002Flanguage, framework, and DB\u002Fbrowser versions from Step 0 (`nvm`\u002F`asdf`\u002F`pyenv`, the committed lockfile, pinned container tags).\n- **Replay data:** load the exact dataset, user state, or recorded API payloads from Step 0 (fixtures, seed scripts, captured requests). State-dependent bugs do not reproduce without their state.\n- **Env vars:** export the captured variables (redact secrets in the report).\n- **Services:** stand up state-dependent dependencies (`docker compose up -d` for DB \u002F cache \u002F queue) when the bug needs them.\n\nCapture **every** command as a reproducibility ledger. Redact tokens and credentials before any output enters the report.\n\n### Step 3 — Reproduce & isolate edge cases\n\nRun the offending path with the Step-0 inputs and confirm the same failure. If it does not fail every time, force consistency:\n\n- **Concurrency \u002F race conditions:** loop the trigger to make an intermittent bug reliable, and record the rate. Add logging around shared state to expose thread-blocking or ordering, and use the language's race detector where available (e.g. `go test -race`).\n\n  ```bash\n  for i in $(seq 1 1000); do \u003Ctrigger-command> || echo \"FAIL run $i\"; done | tee repro-loop.log\n  ```\n\n- **Network & timing:** simulate throttled connections or added latency (`tc`\u002F`netem`, a throttling proxy, or injected delays) to surface timing-dependent races that only appear under real network conditions.\n- **Boundary with `\u002Fdetect-flaky`:** that skill diagnoses non-determinism in an **existing** test suite; this skill reproduces a **new** bug. If the investigation turns into \"why is this existing test flaky\", hand off to `\u002Fdetect-flaky`.\n\n### Step 4 — Write the failing regression test\n\n- Detect the runner — `Makefile` → `make test`; `package.json` → `npm`\u002F`pnpm`\u002F`yarn` test; `go.mod` → `go test .\u002F...`; `pyproject.toml` → `pytest`.\n- Write the **minimal** test that triggers the bug, in the appropriate test file **inside the sandbox**. Assert the **correct** expected behavior so the test fails now and will pass once the bug is fixed.\n- Run it and **paste the actual failure output** — this red result is the objective baseline.\n\n**Gate:** you must have a deterministically failing test, **or** a documented reproduction rate for an intermittent bug (e.g. \"fails 7\u002F1000 runs\"). If you cannot make it fail, STOP — report the evidence gathered and exactly what is still missing. Do not write the report as if the bug were reproduced.\n\n### Step 5 — Write the report\n\nGenerate a filename `\u003Ctopic-slug>-\u003CYYYY-MM-DD>.md` in lowercase kebab-case (e.g. `checkout-500-large-cart-2026-06-14.md`). Create `docs\u002Freproductions\u002F` if it doesn't exist. Use this structure:\n\n````markdown\n# Bug Reproduction: \u003CTitle> — \u003CYYYY-MM-DD>\n\n\u003COne-line symptom.>\n\n## Symptom\n\n- **Observed:** \u003Cwhat happens>\n- **Expected:** \u003Cwhat should happen>\n\n## Evidence Gathered\n\nStack trace:\n\n```text\n\u003Ctop frames>\n```\n\n| Aspect                   | Value                         |\n| ------------------------ | ----------------------------- |\n| Inputs \u002F payload         | \u003Cexact trigger>               |\n| Env vars (redacted)      | \u003CKEY=…>                       |\n| OS \u002F runtime             | \u003Ce.g. Ubuntu 22.04 \u002F Node 20> |\n| Framework \u002F DB \u002F browser | \u003Cversions>                    |\n\nLogs \u002F traces: \u003Cpointer or snippet of the path before the throw>\n\n## Code Path\n\n- `\u003Cfile:line>` — \u003Crole in the failure>\n\n## Isolated Environment\n\nSandbox: `\u003Cabsolute path>`\n\n```bash\n\u003Cexact setup commands: worktree, version pins, seed\u002Freplay, env exports, docker compose>\n```\n\nVersion pins: `\u003Cruntime>`, `\u003Cframework>`, `\u003Cdb\u002Fbrowser>`\n\n## Reproduction\n\n```bash\n\u003Ccommands that trigger the failure>\n```\n\nFailure rate: `\u003Cdeterministic | N\u002FM runs>` (intermittent only)\n\n## Failing Test\n\nPath: `\u003Ctest file>`\n\n```\u003Clang>\n\u003Cthe failing test>\n```\n\nRed baseline:\n\n```text\n\u003Cpasted failure output>\n```\n\n## Suspected Root Cause\n\nCandidates (**not** fixed here):\n\n- \u003Chypothesis> — for: \u003Cevidence> \u002F against: \u003Cevidence>\n\n## Next Step\n\n- `\u002Ffix-with-evidence \u003Cbug>` — the failing test above satisfies its Phase-1 Reproduce gate, OR\n- `\u002Fcreate-experiment \u003Cfix idea>` — compare candidate fixes against this red baseline, OR\n- `\u002Fcreate-audit \u003Carea>` — if the bug looks systemic.\n\n## Sandbox Cleanup\n\n```bash\ngit worktree remove .claude\u002Fworktrees\u002Frepro-\u003Cslug>   # or  rm -rf ~\u002Frepro\u002F\u003Cslug>\n```\n````\n\n### Step 6 — Report back\n\nReply with: the doc path, a one-line repro summary (include the failure rate if intermittent), the sandbox path, and the failing-test path. **Do not paste the full document into chat.** Point the user to the Next Step.\n\n## Rules\n\n- All work runs **locally**. No production endpoints, no real credentials, no shared infrastructure. Mock external services or use disposable test accounts.\n- Capture every setup and reproduction command in the report — reproducibility is the bar.\n- Redact tokens, keys, and PII before anything enters the doc.\n- **Do not fix the bug.** This skill stops at the red baseline; the patch is `\u002Ffix-with-evidence`'s job.\n- The bar is a deterministically failing test, or a documented failure rate for intermittent bugs. \"Probably this\" without a failing run is not a reproduction.\n- Leave the doc, the test, and the sandbox untracked for the user.\n- Tables and code blocks over prose. No filler.\n","Reproduce a bug under isolation and the exact conditions that trigger it, then capture a **failing regression test** as the objective baseline to fix against. The output is a dated report under `docs\u002Freproductions\u002F` plus a failing test left in the sandbox.\n\nTrigger: when the user says \"reproduce this bug\", \"I can't reproduce X\", \"make it fail reliably\", \"repro Y\", or invokes `\u002Freproduce-bug` directly. Use it **after** a bug is reported (or surfaced by `\u002Fcreate-audit`) and **before** `\u002Ffix-with-evidence` writes the patch.\n\nArguments: `$ARGUMENTS` — a bug or symptom (e.g. \"checkout 500s when cart has >50 items\", \"intermittent panic in the scheduler\", issue link\u002Fnumber). If empty, ask the user what to reproduce.\n\n## Purpose\n\n`\u002Freproduce-bug` is **not** a fix (it ships no patch — that is `\u002Ffix-with-evidence`), **not** an audit (it doesn't enumerate issues — that is `\u002Fcreate-audit`), and **not** a flaky-test diagnosis of an existing suite (that is `\u002Fdetect-flaky`). It is the **bridge** between a bug report and a fix: gather the evidence, isolate a prod-mirroring environment, and produce a deterministically failing test.\n\nThe deliverable is a reproduction other skills build on. `\u002Ffix-with-evidence` consumes the failing test directly — it **satisfies that skill's Phase-1 Reproduce gate**, so the fix loop can start at Phase 2.\n\n## Steps\n\n### Step 0 — Gather the failure evidence (interactive, blocking)\n\nCollect the trigger conditions, echo them back, and **wait for sign-off** before touching anything. Skip what is already in `$ARGUMENTS` or recent context. If `$ARGUMENTS` is empty, ask what to reproduce.\n\n1. **Stack trace \u002F error** — the full message and the top frames. For a GitHub issue: `gh issue view \u003CN>`.\n2. **Inputs** — the exact user inputs, API payloads, CLI args, or dataset\u002Fuser-state that triggered the failure.\n3. **Environment** — OS, runtime\u002Flanguage version, framework, and DB\u002Fbrowser versions; the relevant env vars (redact secrets).\n4. **Logs & traces** — for web or distributed systems, the application logs \u002F telemetry that trace the execution path **before** the error was thrown.\n\nEcho back the consolidated evidence **and the precise failure condition you will target**. Wait for explicit user sign-off before continuing to Step 1.\n\n### Step 1 — Locate the code path\n\nUse `Grep` + `Read` to find the throwing path and the code that consumes the Step-0 inputs. Cite `file:line` for every claim. **Do not edit anything yet** — this is baseline-only; root cause is recorded later as candidates, not patched here.\n\n### Step 2 — Build the isolated, prod-mirroring sandbox\n\nIsolate the reproduction so nothing touches the main checkout, then mirror production along every axis the bug depends on.\n\n- **Sandbox:** `git fetch origin main` then `git worktree add .claude\u002Fworktrees\u002Frepro-\u003Cslug> -b repro\u002F\u003Cslug> origin\u002Fmain`. Fallback: `~\u002Frepro\u002F\u003Cslug>\u002F` if there is no enclosing git repo.\n- **Pin versions:** match the runtime\u002Flanguage, framework, and DB\u002Fbrowser versions from Step 0 (`nvm`\u002F`asdf`\u002F`pyenv`, the committed lockfile, pinned container tags).\n- **Replay data:** load the exact dataset, user state, or recorded API payloads from Step 0 (fixtures, seed scripts, captured requests). State-dependent bugs do not reproduce without their state.\n- **Env vars:** export the captured variables (redact secrets in the report).\n- **Services:** stand up state-dependent dependencies (`docker compose up -d` for DB \u002F cache \u002F queue) when the bug needs them.\n\nCapture **every** command as a reproducibility ledger. Redact tokens and credentials before any output enters the report.\n\n### Step 3 — Reproduce & isolate edge cases\n\nRun the offending path with the Step-0 inputs and confirm the same failure. If it does not fail every time, force consistency:\n\n- **Concurrency \u002F race conditions:** loop the trigger to make an intermittent bug reliable, and record the rate. Add logging around shared state to expose thread-blocking or ordering, and use the language's race detector where available (e.g. `go test -race`).\n\n  ```bash\n  for i in $(seq 1 1000); do \u003Ctrigger-command> || echo \"FAIL run $i\"; done | tee repro-loop.log\n  ```\n\n- **Network & timing:** simulate throttled connections or added latency (`tc`\u002F`netem`, a throttling proxy, or injected delays) to surface timing-dependent races that only appear under real network conditions.\n- **Boundary with `\u002Fdetect-flaky`:** that skill diagnoses non-determinism in an **existing** test suite; this skill reproduces a **new** bug. If the investigation turns into \"why is this existing test flaky\", hand off to `\u002Fdetect-flaky`.\n\n### Step 4 — Write the failing regression test\n\n- Detect the runner — `Makefile` → `make test`; `package.json` → `npm`\u002F`pnpm`\u002F`yarn` test; `go.mod` → `go test .\u002F...`; `pyproject.toml` → `pytest`.\n- Write the **minimal** test that triggers the bug, in the appropriate test file **inside the sandbox**. Assert the **correct** expected behavior so the test fails now and will pass once the bug is fixed.\n- Run it and **paste the actual failure output** — this red result is the objective baseline.\n\n**Gate:** you must have a deterministically failing test, **or** a documented reproduction rate for an intermittent bug (e.g. \"fails 7\u002F1000 runs\"). If you cannot make it fail, STOP — report the evidence gathered and exactly what is still missing. Do not write the report as if the bug were reproduced.\n\n### Step 5 — Write the report\n\nGenerate a filename `\u003Ctopic-slug>-\u003CYYYY-MM-DD>.md` in lowercase kebab-case (e.g. `checkout-500-large-cart-2026-06-14.md`). Create `docs\u002Freproductions\u002F` if it doesn't exist. Use this structure:\n\n````markdown\n# Bug Reproduction: \u003CTitle> — \u003CYYYY-MM-DD>\n\n\u003COne-line symptom.>\n\n## Symptom\n\n- **Observed:** \u003Cwhat happens>\n- **Expected:** \u003Cwhat should happen>\n\n## Evidence Gathered\n\nStack trace:\n\n```text\n\u003Ctop frames>\n```\n\n| Aspect                   | Value                         |\n| ------------------------ | ----------------------------- |\n| Inputs \u002F payload         | \u003Cexact trigger>               |\n| Env vars (redacted)      | \u003CKEY=…>                       |\n| OS \u002F runtime             | \u003Ce.g. Ubuntu 22.04 \u002F Node 20> |\n| Framework \u002F DB \u002F browser | \u003Cversions>                    |\n\nLogs \u002F traces: \u003Cpointer or snippet of the path before the throw>\n\n## Code Path\n\n- `\u003Cfile:line>` — \u003Crole in the failure>\n\n## Isolated Environment\n\nSandbox: `\u003Cabsolute path>`\n\n```bash\n\u003Cexact setup commands: worktree, version pins, seed\u002Freplay, env exports, docker compose>\n```\n\nVersion pins: `\u003Cruntime>`, `\u003Cframework>`, `\u003Cdb\u002Fbrowser>`\n\n## Reproduction\n\n```bash\n\u003Ccommands that trigger the failure>\n```\n\nFailure rate: `\u003Cdeterministic | N\u002FM runs>` (intermittent only)\n\n## Failing Test\n\nPath: `\u003Ctest file>`\n\n```\u003Clang>\n\u003Cthe failing test>\n```\n\nRed baseline:\n\n```text\n\u003Cpasted failure output>\n```\n\n## Suspected Root Cause\n\nCandidates (**not** fixed here):\n\n- \u003Chypothesis> — for: \u003Cevidence> \u002F against: \u003Cevidence>\n\n## Next Step\n\n- `\u002Ffix-with-evidence \u003Cbug>` — the failing test above satisfies its Phase-1 Reproduce gate, OR\n- `\u002Fcreate-experiment \u003Cfix idea>` — compare candidate fixes against this red baseline, OR\n- `\u002Fcreate-audit \u003Carea>` — if the bug looks systemic.\n\n## Sandbox Cleanup\n\n```bash\ngit worktree remove .claude\u002Fworktrees\u002Frepro-\u003Cslug>   # or  rm -rf ~\u002Frepro\u002F\u003Cslug>\n```\n````\n\n### Step 6 — Report back\n\nReply with: the doc path, a one-line repro summary (include the failure rate if intermittent), the sandbox path, and the failing-test path. **Do not paste the full document into chat.** Point the user to the Next Step.\n\n## Rules\n\n- All work runs **locally**. No production endpoints, no real credentials, no shared infrastructure. Mock external services or use disposable test accounts.\n- Capture every setup and reproduction command in the report — reproducibility is the bar.\n- Redact tokens, keys, and PII before anything enters the doc.\n- **Do not fix the bug.** This skill stops at the red baseline; the patch is `\u002Ffix-with-evidence`'s job.\n- The bar is a deterministically failing test, or a documented failure rate for intermittent bugs. \"Probably this\" without a failing run is not a reproduction.\n- Leave the doc, the test, and the sandbox untracked for the user.\n- Tables and code blocks over prose. No filler.\n",{"id":5,"name":5,"task":49,"type":50,"model":29,"owner":51,"tools":52,"domain":53,"created":54,"updated":54,"version":55,"dotbabel":56,"maturity":61,"platform":62,"description":64,"argument-hint":65},[43,28],"skill","@kaiohenricunha","Read, Grep, Glob, Bash, Write, Edit",[31],1781395200000,"1.0.0",{"compute":57},{"mode":58,"binding":59,"rationale":60,"requirement":35},"dynamic","consumer","Isolation and regression capture must reproduce the actual triggering conditions.","validated",[63],"none","Reproduce a bug in an isolated, production-mirroring sandbox and capture a failing regression test as the objective baseline to fix against. Saves a report to docs\u002Freproductions\u002F. Triggers on: \"reproduce this bug\", \"can't reproduce\", \"make it fail reliably\", \"repro\".\n","[bug or symptom]","main","91c337164997b99943fbcc64e8e362577b67bc81","skills\u002Freproduce-bug","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":74,"prompt":84,"zipAvailable":85},[75,78,81],{"tool":76,"command":77},"npx","npx skills add kaiohenricunha\u002Fdotbabel --skill reproduce-bug",{"tool":79,"command":80},"bunx","bunx skills add kaiohenricunha\u002Fdotbabel --skill reproduce-bug",{"tool":82,"command":83},"pnpm","pnpm dlx skills add kaiohenricunha\u002Fdotbabel --skill reproduce-bug","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fkaiohenricunha\u002Fdotbabel\u002Fblob\u002Fmain\u002Fskills\u002Freproduce-bug\u002FSKILL.md",true,[],[88,117,148,169,189,210],{"skillNo":89,"name":90,"nameZh":6,"description":91,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":92,"categories":93,"keywords":99,"publishedAt":44},"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",[94],{"categoryNo":95,"categoryName":96,"taxonomyType":97,"taxonomyTypeDesc":98},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[90,28,100,101,102,103,104,105,106,107,108,109,110,111,31,112,113,114,115,116],"merge","code","start","access","session","ops","autonomy","releases","permission","full","runtime","scope","hook","true","action","for","scoped",{"skillNo":118,"name":119,"nameZh":6,"description":120,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":121,"categories":122,"keywords":129,"publishedAt":44},"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",[123,126],{"categoryNo":124,"categoryName":125,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",{"categoryNo":127,"categoryName":128,"taxonomyType":97,"taxonomyTypeDesc":98},"01JBSEEDCAT000000000000037","律师与法务",[130,131,28,132,133,134,42,135,136,137,138,139,140,141,142,143,144,145,146,147],"audit","veracity audit","review","data pipeline","pipeline","math","coverage","only","engineer","formula","scoring","time","project","worker","end","at","invoke","when",{"skillNo":149,"name":150,"nameZh":6,"description":151,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":152,"categories":153,"keywords":154,"publishedAt":44},"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",[],[155,156,157,28,132,158,159,130,42,160,40,161,137,162,163,164,165,166,167,168],"spec","validate","validate spec","run","acceptance","work","never","max","done","arch","line","2025","spec-driven","closure",{"skillNo":170,"name":171,"nameZh":6,"description":172,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":173,"categories":174,"keywords":178,"publishedAt":44},"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",[175],{"categoryNo":176,"categoryName":177,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000011","运维与部署",[179,180,181,132,182,43,158,130,183,42,40,184,185,162,35,186,166,187,145,188],"terragrunt","specialist","terragrunt specialist","orchestration","analysis","dependency","hooks","dry","remote","hierarchy",{"skillNo":190,"name":191,"nameZh":6,"description":192,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":193,"categories":194,"keywords":200,"publishedAt":44},"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",[195,196,197],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":176,"categoryName":177,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":198,"categoryName":199,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000015","视觉设计",[201,202,180,28,203,132,204,205,206,43,130,42,40,136,207,208,35,209,166,145],"terraform","terraform specialist","design","workflows","architecture","architecture review","test coverage","workspaces","problem",{"skillNo":211,"name":155,"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":44},"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":21,"taxonomyTypeDesc":22},{"categoryNo":217,"categoryName":218,"taxonomyType":97,"taxonomyTypeDesc":98},"01JBSEEDCAT000000000000032","设计师",[155,203,220,205,221,222,223,39,224,225,226,227,34,228,229,115,230,231,232,233],"component","create","plan","projects","interactive","use","document","specification","opus","technical specification","trade-offs","technical plan","outputs","frontier"]