[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M71DZ2N050FZ3VT03Q2SS2H0":3,"skill-same:01M71DZ2N050FZ3VT03Q2SS2H0":83},{"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":28,"publishedAt":48,"linkOnly":49,"markdownContent":50,"markdownBody":51,"frontMatter":52,"refName":54,"commitSha":55,"skillDirectory":56,"forksCount":57,"repositoryHtmlUrl":58,"repositoryHomepageUrl":6,"ownerHtmlUrl":59,"ownerBlogUrl":60,"ownerDescription":61,"updatedAt":62,"install":63,"resources":76},"01M71DZ2N050FZ3VT03Q2SS2H0","investigate",null,"Root-cause debugging before any fix: scope lock, hypothesis testing, debug report. Triggers: \"debug this\", \"why is this broken\".","indigoai-us\u002Fhq-core","indigoai-us","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F122242446?v=4","01M71DTQ5NGD10GT1Z7E6XKWC5","HQ","01M71DTQ6BKSYYBNC9RAKT3YDK",94,"MIT","https:\u002F\u002Fgithub.com\u002Findigoai-us\u002Fhq-core\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Finvestigate\u002FSKILL.md",[18,23],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000008","测试与质量","SCENARIO","使用场景",{"categoryNo":24,"categoryName":25,"taxonomyType":26,"taxonomyTypeDesc":27},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[5,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47],"debugging","report","read","scripts","why","work","lock","root-cause","Bash","debug","triggers","fix","company","testing","core","scope","broken","mesh","bind","2026-11-10 10:12:30",false,"---\nname: investigate\ndescription: \"Root-cause debugging before any fix: scope lock, hypothesis testing, debug report. Triggers: \\\"debug this\\\", \\\"why is this broken\\\".\"\nallowed-tools: Read, Grep, Glob, Write, Bash(git:*), Bash(qmd:*), Bash(bash core\u002Fscripts\u002Fresolve-company.sh:*), Bash(bash core\u002Fscripts\u002Fwork-mesh-live-bind-trusted.sh:*), AskUserQuestion\n---\n\n# Iron Law Debugging\n\n**THE IRON LAW: No fixes without root cause investigation first.**\n\nYou are a systematic debugger. Your job is to find the root cause, not to guess at fixes. Resist every urge to \"quickly try\" a fix before you understand what's happening. Quick fixes that work by accident are worse than no fix — they hide the real problem.\n\n**Red flags — stop and reconsider if you catch yourself:**\n- Proposing a fix before tracing the data flow\n- Saying \"quick fix for now\"\n- Each fix reveals a new problem (you're treating symptoms, not the cause)\n- Changing code you don't fully understand\n\n## Step 0: Company Anchor + Scope Lock\n\nResolve the company with the shared resolver — never by hand, and never from the first word alone:\n\n```bash\nbash core\u002Fscripts\u002Fresolve-company.sh --prompt \"{the user's full input}\"\n```\n\nIt returns `{\"company\":\"\u003Cslug>\",\"source\":\"session|prompt|none\"}`: the company bound for this session by `\u002Fstartwork` wins, then a whole-token scan of the entire input (longest slug wins), then nothing. If the resolver is unavailable (older HQ install), fall back to matching the first word against `companies\u002Fmanifest.yaml` top-level keys.\n\nIf matched:\n1. Set `{co}` = matched slug, strip from input\n2. **Bind the session** — run `bash core\u002Fscripts\u002Fwork-mesh-live-bind-trusted.sh --company \"{co}\"` now, before any read under `companies\u002F{co}\u002F`. The scope guard blocks company paths until the session is bound to `{co}`.\n3. Load policies: `companies\u002F{co}\u002Fpolicies\u002F` (skip `example-policy.md`)\n4. Scope qmd searches to company collection if available\n\nIf no match: infer company from cwd (`companies\u002F{slug}\u002F` or `repos\u002F{pub|priv}\u002F{name}` → manifest lookup).\n\n**Scope Lock Declaration:**\n\nAfter understanding the symptom (Step 1), declare the scope lock:\n\n```\nSCOPE LOCK: Restricting investigation to {module\u002Fdirectory\u002Farea}.\nWill not edit files outside this scope during investigation.\n```\n\nThe scope lock is conceptual — a discipline constraint, not a file lock. It prevents fix scope creep. If evidence points outside the locked scope, explicitly acknowledge and expand with justification.\n\n## Step 1: Symptom Capture\n\nIf the symptom is clear from the input, proceed directly. Otherwise, one AskUserQuestion:\n\n**Questions (batch into one call, include only what's missing):**\n\n1. **What is the observable symptom?** — Describe the actual behavior, not the assumed cause\n2. **When did this start?** — After which commit, deploy, config change, or dependency update?\n3. **Is it reproducible?** — What are the exact steps? Does it happen every time or intermittently?\n4. **What environment?** — Production\u002Fstaging\u002Flocal? Which user\u002Faccount? Which platform\u002Fbrowser?\n\n**Critical:** Record the symptom as stated. Do not reinterpret or assume causes yet.\n\n## Step 2: Evidence Gathering (Read-Only)\n\nGather context WITHOUT modifying anything:\n\n```bash\n# Recent commits in relevant area\ngit log --oneline -20 -- {affected paths}\n\n# Search for prior incidents or related discussions\nqmd search \"{symptom keywords}\" --json -n 10\n# If company-scoped: add -c {collection}\n\n# Check workspace threads for similar past issues\nqmd search \"{error message or symptom}\" --json -n 5\n\n# Check audit log for recent task events in this area\ngrep \"{relevant identifiers}\" workspace\u002Fmetrics\u002Faudit-log.jsonl | tail -20\n```\n\nAlso:\n- Read error messages, stack traces, log output provided by user\n- Grep for relevant patterns in affected modules\n- Read the code at the failure point — trace the data flow backward from the symptom\n- Check git diff for recent changes in the affected area\n\n**Present findings before proceeding:**\n```\nEvidence gathered:\n- Recent changes: {relevant commits or \"none in affected area\"}\n- Prior incidents: {similar past issues or \"none found\"}\n- Code trace: {key observation about the data flow}\n- Environment: {relevant env details}\n```\n\n## Step 3: Pattern Classification\n\nClassify the bug against known categories. Rate likelihood for each plausible pattern:\n\n```\nPattern                | Likelihood | Evidence For              | Evidence Against\n-----------------------|------------|---------------------------|------------------\nRACE CONDITION         | H\u002FM\u002FL      | {observation}             | {observation}\nNIL PROPAGATION        | H\u002FM\u002FL      | null\u002Fundefined too far    |\nSTATE CORRUPTION       | H\u002FM\u002FL      | unexpected mutation       |\nCONFIG MISMATCH        | H\u002FM\u002FL      | env var wrong\u002Fmissing     |\nINTEGRATION DRIFT      | H\u002FM\u002FL      | external API changed      |\nTYPE MISMATCH          | H\u002FM\u002FL      | TS types ≠ runtime        |\nHOOK ORDERING          | H\u002FM\u002FL      | lifecycle\u002Feffect order     |\nAUTH BOUNDARY           | H\u002FM\u002FL      | permission\u002Fcredential bug |\nCONCURRENT EDIT        | H\u002FM\u002FL      | parallel agent conflict   |\nCOMPANY ISOLATION      | H\u002FM\u002FL      | cross-company data bleed  |\n```\n\nOnly include rows with M or H likelihood. Skip patterns with no supporting evidence.\n\n**Select top 3 patterns** by likelihood to form hypotheses.\n\n## Step 4: Hypothesis Testing (3-Strike Rule)\n\nGenerate 3 hypotheses in priority order (highest likelihood first).\n\nFor each hypothesis:\n\n```\n### Hypothesis {N}: {precise statement}\n\n**Pattern:** {classification from Step 3}\n**Prediction:** If this is correct, then {observable consequence that can be tested}\n**Test:** {specific read-only test — read a file, check a log, verify a value, trace a path}\n```\n\nExecute the test. Record the result:\n- **CONFIRMED** — evidence supports this hypothesis as the root cause\n- **REFUTED** — evidence contradicts this hypothesis\n- **INCONCLUSIVE** — test didn't produce clear evidence either way\n\n**3-Strike Rule:**\n\nIf ALL 3 hypotheses are refuted or inconclusive:\n\n**STOP.** Do not generate Hypothesis 4 automatically. Instead:\n\n1. Surface what you've learned from the failed hypotheses\n2. Identify what information is still missing\n3. Consider expanding the scope lock if evidence points elsewhere\n4. AskUserQuestion to course-correct:\n   - \"All three hypotheses were ruled out. Here's what I've learned: {summary}. To proceed, I need: A) {specific info}. B) Expand scope to {new area}. C) {alternative approach}.\"\n\nIf a hypothesis is CONFIRMED, proceed to Step 5 immediately — do not test remaining hypotheses.\n\n## Step 5: DEBUG REPORT\n\nWrite a structured report to `workspace\u002Freports\u002F{slug}-debug.md`:\n\n```markdown\n# DEBUG REPORT — {symptom summary}\n\n**Date:** {ISO8601}\n**Company:** {co}\n**Scope lock:** {affected module\u002Farea}\n**Investigation depth:** {N} hypotheses tested\n\n## Root Cause\n\n**Status:** CONFIRMED \u002F SUSPECTED \u002F UNKNOWN\n\n{2-3 sentence description of what is happening and why}\n\n## Evidence\n\n| Evidence | Supports | Refutes |\n|---|---|---|\n| {observation} | Hypothesis {N} | — |\n| {observation} | — | Hypothesis {M} |\n\n## Pattern Classification\n\n**Primary:** {RACE CONDITION \u002F NIL PROPAGATION \u002F etc.}\n**Secondary:** {if applicable, otherwise omit}\n\n## What Was Ruled Out\n\n- **Hypothesis 1** ({pattern}): {refuted because}\n- **Hypothesis 2** ({pattern}): {refuted because}\n\n## Recommended Fix\n\n{Only if root cause is CONFIRMED or strongly SUSPECTED}\n\n**Minimal change:** {describe the smallest possible fix — one function, one file if possible}\n**Files to touch:** {specific file paths — no others}\n**Risk of fix:** LOW \u002F MEDIUM — {does this fix have any side effects?}\n**Regression test:** {what test would catch this if it regressed?}\n\n## If Root Cause Unknown\n\n{List remaining unknowns and what information would resolve them}\n{Suggest next investigation steps}\n```\n\n## Step 6: Fix Gate\n\n**If root cause is CONFIRMED:**\n\nPresent the recommended fix from the DEBUG REPORT. AskUserQuestion:\n- **A) Apply fix now** — Apply ONLY the minimal targeted fix described in the report. No scope creep. No \"while we're here\" improvements. Fix the bug, add a regression test if feasible, done.\n- **B) I'll apply manually** — User takes it from here. Investigation complete.\n- **C) Need more investigation** — Something doesn't add up. Return to Step 4 with expanded scope.\n\nIf A: apply the fix, then verify it resolves the symptom. After verification, suggest `\u002Flearn` to capture the pattern as a policy.\n\n**If root cause is SUSPECTED (not confirmed):**\n\nAskUserQuestion:\n- **A) Proceed with fix under uncertainty** — Apply the fix, but flag it as speculative in the commit message\n- **B) Gather more evidence first** — Return to Step 2 with targeted questions\n\n**If root cause is UNKNOWN:**\n\nOutput remaining unknowns from the DEBUG REPORT. Do NOT suggest guessing or trying random fixes. Investigation is complete — the honest answer is \"I don't know yet.\"\n\n---\n\n## Rules\n\n- **Iron Law above all** — never edit code before completing Steps 1-4. No exceptions\n- **Read before write** — Steps 1-4 are entirely read-only. No file modifications until Step 6\n- **Scope lock is sacred** — do not edit files outside the declared scope without explicitly expanding it\n- **3-strike discipline** — if 3 hypotheses fail, stop and ask. Do not spiral\n- **Minimal fix** — when applying a fix, change the minimum code necessary. No refactoring, no cleanup, no \"improvements\"\n- **Company isolation** — respect all manifest boundaries. Never search or read cross-company data for debugging\n- **Report always written** — even if the root cause is UNKNOWN, write the DEBUG REPORT. Investigation history has value\n- **Do NOT use TodoWrite or EnterPlanMode** — this command is a focused investigation, not a project\n- **No Linear sync** — investigation is pre-task. Linear happens when fixes become stories\n","# Iron Law Debugging\n\n**THE IRON LAW: No fixes without root cause investigation first.**\n\nYou are a systematic debugger. Your job is to find the root cause, not to guess at fixes. Resist every urge to \"quickly try\" a fix before you understand what's happening. Quick fixes that work by accident are worse than no fix — they hide the real problem.\n\n**Red flags — stop and reconsider if you catch yourself:**\n- Proposing a fix before tracing the data flow\n- Saying \"quick fix for now\"\n- Each fix reveals a new problem (you're treating symptoms, not the cause)\n- Changing code you don't fully understand\n\n## Step 0: Company Anchor + Scope Lock\n\nResolve the company with the shared resolver — never by hand, and never from the first word alone:\n\n```bash\nbash core\u002Fscripts\u002Fresolve-company.sh --prompt \"{the user's full input}\"\n```\n\nIt returns `{\"company\":\"\u003Cslug>\",\"source\":\"session|prompt|none\"}`: the company bound for this session by `\u002Fstartwork` wins, then a whole-token scan of the entire input (longest slug wins), then nothing. If the resolver is unavailable (older HQ install), fall back to matching the first word against `companies\u002Fmanifest.yaml` top-level keys.\n\nIf matched:\n1. Set `{co}` = matched slug, strip from input\n2. **Bind the session** — run `bash core\u002Fscripts\u002Fwork-mesh-live-bind-trusted.sh --company \"{co}\"` now, before any read under `companies\u002F{co}\u002F`. The scope guard blocks company paths until the session is bound to `{co}`.\n3. Load policies: `companies\u002F{co}\u002Fpolicies\u002F` (skip `example-policy.md`)\n4. Scope qmd searches to company collection if available\n\nIf no match: infer company from cwd (`companies\u002F{slug}\u002F` or `repos\u002F{pub|priv}\u002F{name}` → manifest lookup).\n\n**Scope Lock Declaration:**\n\nAfter understanding the symptom (Step 1), declare the scope lock:\n\n```\nSCOPE LOCK: Restricting investigation to {module\u002Fdirectory\u002Farea}.\nWill not edit files outside this scope during investigation.\n```\n\nThe scope lock is conceptual — a discipline constraint, not a file lock. It prevents fix scope creep. If evidence points outside the locked scope, explicitly acknowledge and expand with justification.\n\n## Step 1: Symptom Capture\n\nIf the symptom is clear from the input, proceed directly. Otherwise, one AskUserQuestion:\n\n**Questions (batch into one call, include only what's missing):**\n\n1. **What is the observable symptom?** — Describe the actual behavior, not the assumed cause\n2. **When did this start?** — After which commit, deploy, config change, or dependency update?\n3. **Is it reproducible?** — What are the exact steps? Does it happen every time or intermittently?\n4. **What environment?** — Production\u002Fstaging\u002Flocal? Which user\u002Faccount? Which platform\u002Fbrowser?\n\n**Critical:** Record the symptom as stated. Do not reinterpret or assume causes yet.\n\n## Step 2: Evidence Gathering (Read-Only)\n\nGather context WITHOUT modifying anything:\n\n```bash\n# Recent commits in relevant area\ngit log --oneline -20 -- {affected paths}\n\n# Search for prior incidents or related discussions\nqmd search \"{symptom keywords}\" --json -n 10\n# If company-scoped: add -c {collection}\n\n# Check workspace threads for similar past issues\nqmd search \"{error message or symptom}\" --json -n 5\n\n# Check audit log for recent task events in this area\ngrep \"{relevant identifiers}\" workspace\u002Fmetrics\u002Faudit-log.jsonl | tail -20\n```\n\nAlso:\n- Read error messages, stack traces, log output provided by user\n- Grep for relevant patterns in affected modules\n- Read the code at the failure point — trace the data flow backward from the symptom\n- Check git diff for recent changes in the affected area\n\n**Present findings before proceeding:**\n```\nEvidence gathered:\n- Recent changes: {relevant commits or \"none in affected area\"}\n- Prior incidents: {similar past issues or \"none found\"}\n- Code trace: {key observation about the data flow}\n- Environment: {relevant env details}\n```\n\n## Step 3: Pattern Classification\n\nClassify the bug against known categories. Rate likelihood for each plausible pattern:\n\n```\nPattern                | Likelihood | Evidence For              | Evidence Against\n-----------------------|------------|---------------------------|------------------\nRACE CONDITION         | H\u002FM\u002FL      | {observation}             | {observation}\nNIL PROPAGATION        | H\u002FM\u002FL      | null\u002Fundefined too far    |\nSTATE CORRUPTION       | H\u002FM\u002FL      | unexpected mutation       |\nCONFIG MISMATCH        | H\u002FM\u002FL      | env var wrong\u002Fmissing     |\nINTEGRATION DRIFT      | H\u002FM\u002FL      | external API changed      |\nTYPE MISMATCH          | H\u002FM\u002FL      | TS types ≠ runtime        |\nHOOK ORDERING          | H\u002FM\u002FL      | lifecycle\u002Feffect order     |\nAUTH BOUNDARY           | H\u002FM\u002FL      | permission\u002Fcredential bug |\nCONCURRENT EDIT        | H\u002FM\u002FL      | parallel agent conflict   |\nCOMPANY ISOLATION      | H\u002FM\u002FL      | cross-company data bleed  |\n```\n\nOnly include rows with M or H likelihood. Skip patterns with no supporting evidence.\n\n**Select top 3 patterns** by likelihood to form hypotheses.\n\n## Step 4: Hypothesis Testing (3-Strike Rule)\n\nGenerate 3 hypotheses in priority order (highest likelihood first).\n\nFor each hypothesis:\n\n```\n### Hypothesis {N}: {precise statement}\n\n**Pattern:** {classification from Step 3}\n**Prediction:** If this is correct, then {observable consequence that can be tested}\n**Test:** {specific read-only test — read a file, check a log, verify a value, trace a path}\n```\n\nExecute the test. Record the result:\n- **CONFIRMED** — evidence supports this hypothesis as the root cause\n- **REFUTED** — evidence contradicts this hypothesis\n- **INCONCLUSIVE** — test didn't produce clear evidence either way\n\n**3-Strike Rule:**\n\nIf ALL 3 hypotheses are refuted or inconclusive:\n\n**STOP.** Do not generate Hypothesis 4 automatically. Instead:\n\n1. Surface what you've learned from the failed hypotheses\n2. Identify what information is still missing\n3. Consider expanding the scope lock if evidence points elsewhere\n4. AskUserQuestion to course-correct:\n   - \"All three hypotheses were ruled out. Here's what I've learned: {summary}. To proceed, I need: A) {specific info}. B) Expand scope to {new area}. C) {alternative approach}.\"\n\nIf a hypothesis is CONFIRMED, proceed to Step 5 immediately — do not test remaining hypotheses.\n\n## Step 5: DEBUG REPORT\n\nWrite a structured report to `workspace\u002Freports\u002F{slug}-debug.md`:\n\n```markdown\n# DEBUG REPORT — {symptom summary}\n\n**Date:** {ISO8601}\n**Company:** {co}\n**Scope lock:** {affected module\u002Farea}\n**Investigation depth:** {N} hypotheses tested\n\n## Root Cause\n\n**Status:** CONFIRMED \u002F SUSPECTED \u002F UNKNOWN\n\n{2-3 sentence description of what is happening and why}\n\n## Evidence\n\n| Evidence | Supports | Refutes |\n|---|---|---|\n| {observation} | Hypothesis {N} | — |\n| {observation} | — | Hypothesis {M} |\n\n## Pattern Classification\n\n**Primary:** {RACE CONDITION \u002F NIL PROPAGATION \u002F etc.}\n**Secondary:** {if applicable, otherwise omit}\n\n## What Was Ruled Out\n\n- **Hypothesis 1** ({pattern}): {refuted because}\n- **Hypothesis 2** ({pattern}): {refuted because}\n\n## Recommended Fix\n\n{Only if root cause is CONFIRMED or strongly SUSPECTED}\n\n**Minimal change:** {describe the smallest possible fix — one function, one file if possible}\n**Files to touch:** {specific file paths — no others}\n**Risk of fix:** LOW \u002F MEDIUM — {does this fix have any side effects?}\n**Regression test:** {what test would catch this if it regressed?}\n\n## If Root Cause Unknown\n\n{List remaining unknowns and what information would resolve them}\n{Suggest next investigation steps}\n```\n\n## Step 6: Fix Gate\n\n**If root cause is CONFIRMED:**\n\nPresent the recommended fix from the DEBUG REPORT. AskUserQuestion:\n- **A) Apply fix now** — Apply ONLY the minimal targeted fix described in the report. No scope creep. No \"while we're here\" improvements. Fix the bug, add a regression test if feasible, done.\n- **B) I'll apply manually** — User takes it from here. Investigation complete.\n- **C) Need more investigation** — Something doesn't add up. Return to Step 4 with expanded scope.\n\nIf A: apply the fix, then verify it resolves the symptom. After verification, suggest `\u002Flearn` to capture the pattern as a policy.\n\n**If root cause is SUSPECTED (not confirmed):**\n\nAskUserQuestion:\n- **A) Proceed with fix under uncertainty** — Apply the fix, but flag it as speculative in the commit message\n- **B) Gather more evidence first** — Return to Step 2 with targeted questions\n\n**If root cause is UNKNOWN:**\n\nOutput remaining unknowns from the DEBUG REPORT. Do NOT suggest guessing or trying random fixes. Investigation is complete — the honest answer is \"I don't know yet.\"\n\n---\n\n## Rules\n\n- **Iron Law above all** — never edit code before completing Steps 1-4. No exceptions\n- **Read before write** — Steps 1-4 are entirely read-only. No file modifications until Step 6\n- **Scope lock is sacred** — do not edit files outside the declared scope without explicitly expanding it\n- **3-strike discipline** — if 3 hypotheses fail, stop and ask. Do not spiral\n- **Minimal fix** — when applying a fix, change the minimum code necessary. No refactoring, no cleanup, no \"improvements\"\n- **Company isolation** — respect all manifest boundaries. Never search or read cross-company data for debugging\n- **Report always written** — even if the root cause is UNKNOWN, write the DEBUG REPORT. Investigation history has value\n- **Do NOT use TodoWrite or EnterPlanMode** — this command is a focused investigation, not a project\n- **No Linear sync** — investigation is pre-task. Linear happens when fixes become stories\n",{"name":5,"description":7,"allowed-tools":53},"Read, Grep, Glob, Write, Bash(git:*), Bash(qmd:*), Bash(bash core\u002Fscripts\u002Fresolve-company.sh:*), Bash(bash core\u002Fscripts\u002Fwork-mesh-live-bind-trusted.sh:*), AskUserQuestion","main","205cf61728a27ef15e9153aa38df5856fc8d82f5",".claude\u002Fskills\u002Finvestigate",18,"https:\u002F\u002Fgithub.com\u002Findigoai-us\u002Fhq-core","https:\u002F\u002Fgithub.com\u002Findigoai-us","https:\u002F\u002Fhqforwork.com\u002F","The AI Context Layer for Teams","2026-11-09 23:36:37",{"source":8,"skillName":5,"commands":64,"prompt":74,"zipAvailable":75},[65,68,71],{"tool":66,"command":67},"npx","npx skills add indigoai-us\u002Fhq-core --skill investigate",{"tool":69,"command":70},"bunx","bunx skills add indigoai-us\u002Fhq-core --skill investigate",{"tool":72,"command":73},"pnpm","pnpm dlx skills add indigoai-us\u002Fhq-core --skill investigate","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Findigoai-us\u002Fhq-core\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Finvestigate\u002FSKILL.md",true,[77],{"path":78,"resourceType":79,"resourceTypeDesc":80,"sizeBytes":81,"sourceHtmlUrl":82},"agents\u002Fopenai.yaml","REFERENCE","参考文档",117,"https:\u002F\u002Fgithub.com\u002Findigoai-us\u002Fhq-core\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Finvestigate\u002Fagents\u002Fopenai.yaml",[84,104,129,151,162,172],{"skillNo":85,"name":86,"nameZh":6,"description":87,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":88,"categories":89,"keywords":90,"publishedAt":48},"01M71E4MQ04WPK1EVDS7VKJ3E8","social-graphic","Generate 1080x1080 HTML social graphics for HQ update posts.","https:\u002F\u002Fgithub.com\u002Findigoai-us\u002Fhq-core\u002Fblob\u002Fmain\u002Fcore\u002Fworkers\u002Fpublic\u002Fsocial-strategist\u002Fskills\u002Fsocial-graphic\u002FSKILL.md",[],[91,92,37,93,94,95,31,96,97,98,99,100,101,102,103],"social","social graphic","html","edit","write","update","generate","graphics","browser","open","agent-browser","for","posts",{"skillNo":105,"name":101,"nameZh":6,"description":106,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":107,"categories":108,"keywords":113,"publishedAt":48},"01M71E4H33K4P5H5M7ENPTMFSC","Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to \"open a website\", \"fill out a form\", \"click a button\", \"take a screenshot\", \"scrape data from a page\", \"test this web app\", \"login to a site\", \"automate browser actions\", or any task requiring programmatic web interaction.","https:\u002F\u002Fgithub.com\u002Findigoai-us\u002Fhq-core\u002Fblob\u002Fmain\u002Fcore\u002Fworkers\u002Fpublic\u002Fqa-tester\u002Fskills\u002Fagent-browser\u002FSKILL.md",[109,110],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":111,"categoryName":112,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",[114,99,115,116,117,118,37,119,120,121,122,123,100,42,101,124,125,126,127,128],"agent browser","click","ai agents","app","scrape","when","apps","actions","extracting","take a screenshot","programmatic","browser automation","website","cli","interaction",{"skillNo":130,"name":131,"nameZh":6,"description":132,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":133,"categories":134,"keywords":135,"publishedAt":48},"01M71E4FDBD7J8Y45JJ9KMMYGN","work-mesh","Work Mesh Live — automatic presence via hq mesh daemon; context reconcile\u002Forganize; manual session task-status, blocked, and note only.","https:\u002F\u002Fgithub.com\u002Findigoai-us\u002Fhq-core\u002Fblob\u002Fmain\u002Fcore\u002Fskills\u002Fwork-mesh\u002FSKILL.md",[],[136,34,46,37,137,138,139,140,141,142,143,144,145,146,31,147,148,149,150],"work mesh","status","context","note","live","presence","automatic","reconcile","only","organize","task","session","manual","daemon","via",{"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":48},"01M71E4EGHNAAGWAPV6KGBMQ9W","review","Parity worker review skill (materialization fixture).","https:\u002F\u002Fgithub.com\u002Findigoai-us\u002Fhq-core\u002Fblob\u002Fmain\u002Fcore\u002Fscripts\u002Ftests\u002Ffixtures\u002Fagent-session\u002Fworker-materialization\u002Fparity-worker\u002Fskills\u002Freview\u002FSKILL.md",[],[153,158,159,160,161],"review skill","fixture","parity","worker",{"skillNo":163,"name":164,"nameZh":6,"description":165,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":166,"categories":167,"keywords":168,"publishedAt":48},"01M71E4D1FZDJPDC15DRERKX8Y","run","Run or list HQ workers (parity fixture).","https:\u002F\u002Fgithub.com\u002Findigoai-us\u002Fhq-core\u002Fblob\u002Fmain\u002Fcore\u002Fscripts\u002Ftests\u002Ffixtures\u002Fagent-session\u002Fskills\u002Frun\u002FSKILL.md",[],[164,169,170,160,171,159],"parity fixture","workers","list",{"skillNo":173,"name":174,"nameZh":6,"description":175,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":176,"categories":177,"keywords":178,"publishedAt":48},"01M71E4BF4MV5SEWZJZ4T1QPQW","handoff","Preserve session state for a follow-up agent (parity fixture).","https:\u002F\u002Fgithub.com\u002Findigoai-us\u002Fhq-core\u002Fblob\u002Fmain\u002Fcore\u002Fscripts\u002Ftests\u002Ffixtures\u002Fagent-session\u002Fskills\u002Fhandoff\u002FSKILL.md",[],[174,169,179,180,147,181,102,160,182,183,159],"up","preserve","session state","follow-up","state"]