[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6NJ2VP4ECDX78R8K7DR9ZV9":3,"skill-same:01M6NJ2VP4ECDX78R8K7DR9ZV9":121},{"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":40,"linkOnly":41,"markdownContent":42,"markdownBody":43,"frontMatter":44,"refName":49,"commitSha":50,"skillDirectory":51,"forksCount":52,"repositoryHtmlUrl":53,"repositoryHomepageUrl":54,"ownerHtmlUrl":55,"ownerBlogUrl":56,"ownerDescription":57,"updatedAt":58,"install":59,"resources":72},"01M6NJ2VP4ECDX78R8K7DR9ZV9","hunt",null,"Find the root cause before any fix: build a red-capable repro loop, rank falsifiable hypotheses, prove, fix, sweep siblings. Also triages bug issues\u002FPRs. Use on errors, crashes, regressions.","tinhtute\u002Fmonoharness","tinhtute","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F51282340?v=4","01M6NHYENXPR3K033S3YDA62FW","TINHTUTE","01M6NHYETDANVFY82EPKRC4JFS",5,"MIT","https:\u002F\u002Fgithub.com\u002Ftinhtute\u002Fmonoharness\u002Fblob\u002Fmaster\u002Fskills\u002Freason\u002Fhunt\u002FSKILL.md",[],[5,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39],"repro","root cause","prove","prs","for","errors","error","build","issue","bug","diagnose","use","fix","find","sweep","regression","loop","bisect","issues","code","pr","2026-11-10 10:12:30",false,"---\nname: hunt\nversion: \"1.1.0\"\nmodel: opus\ndescription: \"Find the root cause before any fix: build a red-capable repro loop, rank falsifiable hypotheses, prove, fix, sweep siblings. Also triages bug issues\u002FPRs. Use on errors, crashes, regressions.\"\nargument-hint: \"[diagnose|fix|bisect|regression|sweep|triage] [symptom, error, issue\u002FPR ref]\"\ncompatibility: Designed for Claude Code\n---\n\n# Hunt: Diagnose Before You Fix\n\n## Outcome Contract\n\n- Outcome: the root cause is proven before any fix.\n- Done when: one sentence explains the cause, every observed symptom fits it, and the fix (or handoff) is verified by the same loop that reproduced the bug.\n- Evidence: loop command and output, source trace, logs or state, targeted test\u002Fbuild output, runtime evidence for UI\u002Fnative defects.\n- Authorization: \"diagnose\", \"investigate\", \"why\", \"look into\", \"debug\", \"xem thử\", \"sao lỗi\" is report-only. Edit code only on a request to fix, change, or implement, or while that authorization holds for the same unfinished task; never covers commit, push, publish, or anything destructive.\n\n**Do not touch code until you can write:**\n> \"I believe the root cause is [X at file:line \u002F condition] because [evidence].\"\n\nNot \"a state management issue\" but \"stale cache in `useUser` at `src\u002Fhooks\u002Fuser.ts:42` because the dependency array omits `userId`\". Less specific → no hypothesis yet.\n\n## Modes\n\nResolve from a leading subcommand, else the request.\n\n| Mode | Activate when | Does |\n|---|---|---|\n| `diagnose` (default) | Error, crash, test failure, \"not working\", \"why\" | Steps 1–4, report only |\n| `fix` | \"fix it\", \"sửa đi\" | Full loop, steps 1–7 |\n| `bisect` | \"used to work\", \"broke after update\", a known-good commit or version | Step 1 is a bisect harness; read `references\u002Fbisect-regression.md` |\n| `regression` | Same issue after a fix, or a \"good\" screenshot\u002Fversion\u002Ffile to compare | The reference is the oracle; read `references\u002Fbisect-regression.md` |\n| `sweep` | After a root-cause fix, \"anywhere else like this?\", \"còn chỗ nào giống vậy không\" | Step 6 only, on a named pattern; read `references\u002Ffix.md` |\n| `triage` | Issue\u002FPR queue, \"look at #42\", \"what needs my attention\", label\u002Fclose\u002Fbrief | `references\u002Ftriage.md` replaces steps 2–7; its bug check reuses step 1 |\n\n## The Loop\n\n### 1. Build a feedback loop (this is the skill)\n\nSpend most effort here; load `references\u002Ffeedback-loops.md` for ten loop recipes, tightening, and flaky bugs.\n\nExit criterion: **one command already run**, shown with (redacted) output, that is:\n\n- Red-capable: drives the real code path and asserts the user's exact symptom, not \"didn't crash\".\n- Deterministic: same verdict every run (flaky: a pinned, high reproduction rate).\n- Fast (seconds).\n- Agent-runnable: unattended; a human only through `scripts\u002Fhitl-loop.template.sh`.\n\nNo theory from reading code before this command exists. No loop possible → stop, list attempts, ask for environment access, a redacted artifact (HAR, log, core dump, recording), or temporary instrumentation. Only the reporter has the environment → hand them a read-only probe script (prints env, the disputed value, the state the hypothesis turns on; nothing secret): one command, one block pasted back.\n\n### 2. Reproduce and minimise\n\n- The loop must show the **user's** failure, not a nearby one.\n- List every symptom in the user's words, including \"unrelated\" ones.\n- Cut inputs, callers, config, data, and steps one at a time, re-running, until every remaining one is load-bearing.\n\n### 3. Hypothesise: 3–5 ranked, falsifiable\n\nEach states its prediction:\n> \"If X is the cause, then changing Y makes the bug disappear \u002F changing Z makes it worse.\"\n\nNo prediction → sharpen or discard. Covers only some symptoms → symptom-level guess. Show the ranked list to the user before testing if present; do not block on it.\n\n### 4. Instrument, then confirm or discard\n\n- Every probe maps to one prediction; one variable at a time.\n- Debugger\u002FREPL first, then targeted logs at hypothesis boundaries; tag every temporary log with a unique prefix (`[DEBUG-a4f2]`); never \"log everything and grep\"; playbook and per-bug-class tactics: `references\u002Flogging-techniques.md`.\n- Run the probe that would fail if the hypothesis were wrong. Contradicted → discard it and re-orient on what the probe showed. A log that changes the behavior is timing\u002Flifecycle evidence, not noise.\n\nFor a recurring or runtime-state symptom (caches, queues, generated output, PATH, locale, timeouts, entry points), read `## Gotchas` in `references\u002Ffailure-patterns.md`, list its headings (`grep -n '^## '`), and read only the section matching the symptom before adding a second fix. Domain traps: `references\u002Fime-unicode.md` (IME, cursor drift, emoji), `references\u002Frendering-debug.md` (PDF, print, fonts).\n\n### 5–7. Fix, sweep, clean up\n\n`fix`, or `bisect`\u002F`regression` with fix authorization: read `references\u002Ffix.md` and run steps 5–7.\n\n## Hard Rules\n\n- **Same symptom after a fix: hard stop.** Re-read the execution path before touching code.\n- **Three failed hypotheses → stop**, report with the Handoff format in `references\u002Foutput.md`, ask how to proceed.\n- **Measure the lower layer first** for system or tooling symptoms: OS capture vs post-processing, service vs UI, toolchain vs assertion, network vs client handling.\n- **External tool failure: diagnose before switching** (server up, key valid, config right).\n- **Magic number tuned three times: stop.** Unify into one named token; find the missing constraint.\n- **Redact every secret** in commands, outputs, and artifacts you show (`\u003CREDACTED>`); build loops against env vars.\n- **Fix the cause inside the authorized scope.** A prerequisite refactor is fine if you state why; unrequested features and unrelated refactors are not.\n\n## Output\n\nReport with the block and status in `references\u002Foutput.md`; blocked → its Handoff format.\n\nDefer to: `check` (diff review, no symptom), `think` (is it worth it), `brainstorm` (product decision), `git` (commit).\n","# Hunt: Diagnose Before You Fix\n\n## Outcome Contract\n\n- Outcome: the root cause is proven before any fix.\n- Done when: one sentence explains the cause, every observed symptom fits it, and the fix (or handoff) is verified by the same loop that reproduced the bug.\n- Evidence: loop command and output, source trace, logs or state, targeted test\u002Fbuild output, runtime evidence for UI\u002Fnative defects.\n- Authorization: \"diagnose\", \"investigate\", \"why\", \"look into\", \"debug\", \"xem thử\", \"sao lỗi\" is report-only. Edit code only on a request to fix, change, or implement, or while that authorization holds for the same unfinished task; never covers commit, push, publish, or anything destructive.\n\n**Do not touch code until you can write:**\n> \"I believe the root cause is [X at file:line \u002F condition] because [evidence].\"\n\nNot \"a state management issue\" but \"stale cache in `useUser` at `src\u002Fhooks\u002Fuser.ts:42` because the dependency array omits `userId`\". Less specific → no hypothesis yet.\n\n## Modes\n\nResolve from a leading subcommand, else the request.\n\n| Mode | Activate when | Does |\n|---|---|---|\n| `diagnose` (default) | Error, crash, test failure, \"not working\", \"why\" | Steps 1–4, report only |\n| `fix` | \"fix it\", \"sửa đi\" | Full loop, steps 1–7 |\n| `bisect` | \"used to work\", \"broke after update\", a known-good commit or version | Step 1 is a bisect harness; read `references\u002Fbisect-regression.md` |\n| `regression` | Same issue after a fix, or a \"good\" screenshot\u002Fversion\u002Ffile to compare | The reference is the oracle; read `references\u002Fbisect-regression.md` |\n| `sweep` | After a root-cause fix, \"anywhere else like this?\", \"còn chỗ nào giống vậy không\" | Step 6 only, on a named pattern; read `references\u002Ffix.md` |\n| `triage` | Issue\u002FPR queue, \"look at #42\", \"what needs my attention\", label\u002Fclose\u002Fbrief | `references\u002Ftriage.md` replaces steps 2–7; its bug check reuses step 1 |\n\n## The Loop\n\n### 1. Build a feedback loop (this is the skill)\n\nSpend most effort here; load `references\u002Ffeedback-loops.md` for ten loop recipes, tightening, and flaky bugs.\n\nExit criterion: **one command already run**, shown with (redacted) output, that is:\n\n- Red-capable: drives the real code path and asserts the user's exact symptom, not \"didn't crash\".\n- Deterministic: same verdict every run (flaky: a pinned, high reproduction rate).\n- Fast (seconds).\n- Agent-runnable: unattended; a human only through `scripts\u002Fhitl-loop.template.sh`.\n\nNo theory from reading code before this command exists. No loop possible → stop, list attempts, ask for environment access, a redacted artifact (HAR, log, core dump, recording), or temporary instrumentation. Only the reporter has the environment → hand them a read-only probe script (prints env, the disputed value, the state the hypothesis turns on; nothing secret): one command, one block pasted back.\n\n### 2. Reproduce and minimise\n\n- The loop must show the **user's** failure, not a nearby one.\n- List every symptom in the user's words, including \"unrelated\" ones.\n- Cut inputs, callers, config, data, and steps one at a time, re-running, until every remaining one is load-bearing.\n\n### 3. Hypothesise: 3–5 ranked, falsifiable\n\nEach states its prediction:\n> \"If X is the cause, then changing Y makes the bug disappear \u002F changing Z makes it worse.\"\n\nNo prediction → sharpen or discard. Covers only some symptoms → symptom-level guess. Show the ranked list to the user before testing if present; do not block on it.\n\n### 4. Instrument, then confirm or discard\n\n- Every probe maps to one prediction; one variable at a time.\n- Debugger\u002FREPL first, then targeted logs at hypothesis boundaries; tag every temporary log with a unique prefix (`[DEBUG-a4f2]`); never \"log everything and grep\"; playbook and per-bug-class tactics: `references\u002Flogging-techniques.md`.\n- Run the probe that would fail if the hypothesis were wrong. Contradicted → discard it and re-orient on what the probe showed. A log that changes the behavior is timing\u002Flifecycle evidence, not noise.\n\nFor a recurring or runtime-state symptom (caches, queues, generated output, PATH, locale, timeouts, entry points), read `## Gotchas` in `references\u002Ffailure-patterns.md`, list its headings (`grep -n '^## '`), and read only the section matching the symptom before adding a second fix. Domain traps: `references\u002Fime-unicode.md` (IME, cursor drift, emoji), `references\u002Frendering-debug.md` (PDF, print, fonts).\n\n### 5–7. Fix, sweep, clean up\n\n`fix`, or `bisect`\u002F`regression` with fix authorization: read `references\u002Ffix.md` and run steps 5–7.\n\n## Hard Rules\n\n- **Same symptom after a fix: hard stop.** Re-read the execution path before touching code.\n- **Three failed hypotheses → stop**, report with the Handoff format in `references\u002Foutput.md`, ask how to proceed.\n- **Measure the lower layer first** for system or tooling symptoms: OS capture vs post-processing, service vs UI, toolchain vs assertion, network vs client handling.\n- **External tool failure: diagnose before switching** (server up, key valid, config right).\n- **Magic number tuned three times: stop.** Unify into one named token; find the missing constraint.\n- **Redact every secret** in commands, outputs, and artifacts you show (`\u003CREDACTED>`); build loops against env vars.\n- **Fix the cause inside the authorized scope.** A prerequisite refactor is fine if you state why; unrequested features and unrelated refactors are not.\n\n## Output\n\nReport with the block and status in `references\u002Foutput.md`; blocked → its Handoff format.\n\nDefer to: `check` (diff review, no symptom), `think` (is it worth it), `brainstorm` (product decision), `git` (commit).\n",{"name":5,"model":45,"version":46,"description":7,"argument-hint":47,"compatibility":48},"opus","1.1.0","[diagnose|fix|bisect|regression|sweep|triage] [symptom, error, issue\u002FPR ref]","Designed for Claude Code","master","00bd5271da30679f867f003d9d0b09ee174a7152","skills\u002Freason\u002Fhunt",0,"https:\u002F\u002Fgithub.com\u002Ftinhtute\u002Fmonoharness","https:\u002F\u002Fharness.tinhtute.me","https:\u002F\u002Fgithub.com\u002Ftinhtute","tinhtute.me","𝙻 𝙾 𝚂 𝙴 𝚁 ✨🚀","2026-11-11 23:47:07",{"source":8,"skillName":5,"commands":60,"prompt":70,"zipAvailable":71},[61,64,67],{"tool":62,"command":63},"npx","npx skills add tinhtute\u002Fmonoharness --skill hunt",{"tool":65,"command":66},"bunx","bunx skills add tinhtute\u002Fmonoharness --skill hunt",{"tool":68,"command":69},"pnpm","pnpm dlx skills add tinhtute\u002Fmonoharness --skill hunt","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Ftinhtute\u002Fmonoharness\u002Fblob\u002Fmaster\u002Fskills\u002Freason\u002Fhunt\u002FSKILL.md",true,[73,79,83,87,91,95,99,103,107,111,115],{"path":74,"resourceType":75,"resourceTypeDesc":76,"sizeBytes":77,"sourceHtmlUrl":78},"NOTICE.md","REFERENCE","参考文档",205,"https:\u002F\u002Fgithub.com\u002Ftinhtute\u002Fmonoharness\u002Fblob\u002Fmaster\u002Fskills\u002Freason\u002Fhunt\u002FNOTICE.md",{"path":80,"resourceType":75,"resourceTypeDesc":76,"sizeBytes":81,"sourceHtmlUrl":82},"references\u002Fbisect-regression.md",959,"https:\u002F\u002Fgithub.com\u002Ftinhtute\u002Fmonoharness\u002Fblob\u002Fmaster\u002Fskills\u002Freason\u002Fhunt\u002Freferences\u002Fbisect-regression.md",{"path":84,"resourceType":75,"resourceTypeDesc":76,"sizeBytes":85,"sourceHtmlUrl":86},"references\u002Ffailure-patterns.md",16211,"https:\u002F\u002Fgithub.com\u002Ftinhtute\u002Fmonoharness\u002Fblob\u002Fmaster\u002Fskills\u002Freason\u002Fhunt\u002Freferences\u002Ffailure-patterns.md",{"path":88,"resourceType":75,"resourceTypeDesc":76,"sizeBytes":89,"sourceHtmlUrl":90},"references\u002Ffeedback-loops.md",2924,"https:\u002F\u002Fgithub.com\u002Ftinhtute\u002Fmonoharness\u002Fblob\u002Fmaster\u002Fskills\u002Freason\u002Fhunt\u002Freferences\u002Ffeedback-loops.md",{"path":92,"resourceType":75,"resourceTypeDesc":76,"sizeBytes":93,"sourceHtmlUrl":94},"references\u002Ffix.md",2163,"https:\u002F\u002Fgithub.com\u002Ftinhtute\u002Fmonoharness\u002Fblob\u002Fmaster\u002Fskills\u002Freason\u002Fhunt\u002Freferences\u002Ffix.md",{"path":96,"resourceType":75,"resourceTypeDesc":76,"sizeBytes":97,"sourceHtmlUrl":98},"references\u002Fime-unicode.md",4671,"https:\u002F\u002Fgithub.com\u002Ftinhtute\u002Fmonoharness\u002Fblob\u002Fmaster\u002Fskills\u002Freason\u002Fhunt\u002Freferences\u002Fime-unicode.md",{"path":100,"resourceType":75,"resourceTypeDesc":76,"sizeBytes":101,"sourceHtmlUrl":102},"references\u002Flogging-techniques.md",4908,"https:\u002F\u002Fgithub.com\u002Ftinhtute\u002Fmonoharness\u002Fblob\u002Fmaster\u002Fskills\u002Freason\u002Fhunt\u002Freferences\u002Flogging-techniques.md",{"path":104,"resourceType":75,"resourceTypeDesc":76,"sizeBytes":105,"sourceHtmlUrl":106},"references\u002Foutput.md",852,"https:\u002F\u002Fgithub.com\u002Ftinhtute\u002Fmonoharness\u002Fblob\u002Fmaster\u002Fskills\u002Freason\u002Fhunt\u002Freferences\u002Foutput.md",{"path":108,"resourceType":75,"resourceTypeDesc":76,"sizeBytes":109,"sourceHtmlUrl":110},"references\u002Frendering-debug.md",1356,"https:\u002F\u002Fgithub.com\u002Ftinhtute\u002Fmonoharness\u002Fblob\u002Fmaster\u002Fskills\u002Freason\u002Fhunt\u002Freferences\u002Frendering-debug.md",{"path":112,"resourceType":75,"resourceTypeDesc":76,"sizeBytes":113,"sourceHtmlUrl":114},"references\u002Ftriage.md",6455,"https:\u002F\u002Fgithub.com\u002Ftinhtute\u002Fmonoharness\u002Fblob\u002Fmaster\u002Fskills\u002Freason\u002Fhunt\u002Freferences\u002Ftriage.md",{"path":116,"resourceType":117,"resourceTypeDesc":118,"sizeBytes":119,"sourceHtmlUrl":120},"scripts\u002Fhitl-loop.template.sh","SCRIPT","脚本",1316,"https:\u002F\u002Fgithub.com\u002Ftinhtute\u002Fmonoharness\u002Fblob\u002Fmaster\u002Fskills\u002Freason\u002Fhunt\u002Fscripts\u002Fhitl-loop.template.sh",[122,151,177,207,227,249],{"skillNo":123,"name":124,"nameZh":6,"description":125,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":126,"categories":127,"keywords":133,"publishedAt":40},"01M6NJ7YY1Q5V2JYZ401YHKJVK","read","Search the web and fetch URLs or PDFs as clean Markdown via a privacy-first cascade (local, gh, TinyFish\u002FFirecrawl\u002FExa, opt-in proxy). Use to read, cite or save a URL. Not for repo files.","https:\u002F\u002Fgithub.com\u002Ftinhtute\u002Fmonoharness\u002Fblob\u002Fmaster\u002Fskills\u002Ftools\u002Fread\u002FSKILL.md",[128],{"categoryNo":129,"categoryName":130,"taxonomyType":131,"taxonomyTypeDesc":132},"01JBSEEDCAT000000000000002","文档处理","SCENARIO","使用场景",[124,134,135,38,136,137,138,139,140,141,30,142,143,144,145,146,147,148,149,23,150],"PDF","web","search","proxy","first","gh","repo","fetch","files","clean","exa","tinyfish","query","local","as","URL","opt",{"skillNo":152,"name":153,"nameZh":6,"description":154,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":155,"categories":156,"keywords":157,"publishedAt":40},"01M6WWM7Y4Q4C1NN9DRC33J6F8","mono-write","Polish a research or learn draft (EN\u002FVI) so it reads naturally, and write CHANGELOG entries and release notes from a commit list. Not for UI copy, marketing or tone changes.","https:\u002F\u002Fgithub.com\u002Ftinhtute\u002Fmonoharness\u002Fblob\u002Fmaster\u002Fskills\u002Ftools\u002Fmono-write\u002FSKILL.md",[],[158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,23,176],"write","mono","mono write","research","write changelog","commit","notes","release notes","learn","release","changelog","polish","marketing","list","en","draft","entries","tone","changes",{"skillNo":178,"name":179,"nameZh":6,"description":180,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":181,"categories":182,"keywords":191,"publishedAt":40},"01M6NJ76P0AMQBPZSZQB38P0TT","librarian","GitHub code research via gh CLI, no cloning. Use to investigate external repos, find where a symbol is defined, or gather cited evidence and usage examples from GitHub.","https:\u002F\u002Fgithub.com\u002Ftinhtute\u002Fmonoharness\u002Fblob\u002Fmaster\u002Fskills\u002Ftools\u002Flibrarian\u002FSKILL.md",[183,186],{"categoryNo":184,"categoryName":185,"taxonomyType":131,"taxonomyTypeDesc":132},"01JBSEEDCAT000000000000021","资料检索与研究",{"categoryNo":187,"categoryName":188,"taxonomyType":189,"taxonomyTypeDesc":190},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[179,192,193,161,194,38,136,195,196,197,139,198,199,140,124,30,200,201,202,146,203,204,23,205,206],"cli","github","repos","examples","investigate","Bash","haiku","gh cli","evidence","cloning","code research","github code","usage","gh-cli","gather",{"skillNo":208,"name":209,"nameZh":6,"description":210,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":211,"categories":212,"keywords":213,"publishedAt":40},"01M6NJ72XJYEV4VDYBZVT69E5T","git","Git operations with conventional commits. Use for staging, committing, pushing, PRs, merges. Auto-splits commits by type\u002Fscope. Security scans for secrets.","https:\u002F\u002Fgithub.com\u002Ftinhtute\u002Fmonoharness\u002Fblob\u002Fmaster\u002Fskills\u002Ftools\u002Fgit\u002FSKILL.md",[],[209,214,39,38,215,216,198,217,30,218,219,220,221,222,223,224,225,23,226,22],"merge","secrets","commits","operations","auto","cm","conventional commits","type","cp","git operations","scope","with","splits",{"skillNo":228,"name":229,"nameZh":6,"description":230,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":231,"categories":232,"keywords":233,"publishedAt":40},"01M6NJ6BBWY0ZTGKE4TNA6W3KF","work","Execution orchestrator after `brainstorm` (and `to-plan` for high-risk). Runs phases wave-by-wave from locked artifacts, verifies each task, and hands off to `check`, `git`, or `handoff`.","https:\u002F\u002Fgithub.com\u002Ftinhtute\u002Fmonoharness\u002Fblob\u002Fmaster\u002Fskills\u002Fstages\u002Fwork\u002FSKILL.md",[],[229,234,164,38,209,235,236,237,238,239,240,241,218,242,243,244,245,246,247,23,248],"handoff","artifacts","execution","risk","plan","brainstorm","check","orchestrator","simple","task","full","name","mode","high","phase",{"skillNo":250,"name":251,"nameZh":6,"description":252,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":253,"categories":254,"keywords":255,"publishedAt":40},"01M6WWM38JH25Y72M733S1WT9G","wayfinder","Chart a Goal too fogged for one session as a decision map in the active plan, settle one decision per session, then hand it to brainstorm lock. Use when grilling shows the work spans sessions.","https:\u002F\u002Fgithub.com\u002Ftinhtute\u002Fmonoharness\u002Fblob\u002Fmaster\u002Fskills\u002Fstages\u002Fwayfinder\u002FSKILL.md",[],[251,256,38,257,238,258,239,229,259,30,260,261,148,262,263,264,265,23,266,267],"sessions","file","decision","session","resume","map","chart","hand","lock","decision map","when","settle"]