[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7AX94GPNZ2M2Q5AMQAZ7S1W":3,"skill-same:01M7AX94GPNZ2M2Q5AMQAZ7S1W":47},{"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":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":24,"commitSha":25,"skillDirectory":26,"forksCount":27,"repositoryHtmlUrl":28,"repositoryHomepageUrl":6,"ownerHtmlUrl":29,"ownerBlogUrl":30,"ownerDescription":31,"updatedAt":32,"install":33,"resources":46},"01M7AX94GPNZ2M2Q5AMQAZ7S1W","community-triage",null,"Triage community issues, pull requests, discussions, and support threads against project scope and maintainer capacity. Use when reading or replying to a GitHub issue or PR, deciding whether a request is in scope, judging whether a report carries enough evidence, routing a limitation to a harness vendor, or closing a thread.","roryeckel\u002Fhooknostic","roryeckel","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F16675082?v=4","01M7AX8JWEFHND7P2G338X0TQW","Rory","01M7AX8JX2FGW3TMBVSGFAPQD2",2,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Froryeckel\u002Fhooknostic\u002Fblob\u002Fmaster\u002F.agents\u002Fskills\u002Fcommunity-triage\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: community-triage\ndescription: Triage community issues, pull requests, discussions, and support threads against project scope and maintainer capacity. Use when reading or replying to a GitHub issue or PR, deciding whether a request is in scope, judging whether a report carries enough evidence, routing a limitation to a harness vendor, or closing a thread.\n---\n\n# Community triage\n\nYou are triaging for a **single-maintainer** project whose product is honest\ncapability information, not maximal coverage. Most threads are decided by scope\nand evidence rather than by effort, and saying so kindly and early is the\nservice — a request left open for months is worse for the reporter than a clear\n\"not this project, and here is why\".\n\n## Required context\n\nThe canonical scope statement lives in `CONTRIBUTING.md` (\"What Hooknostic is\" \u002F\n\"What Hooknostic is not\" \u002F \"The rule that decides most contributions\"). Read it\nand **link to it**; do not restate it in your reply, and never let this file\ndrift into a second copy of it.\n\nFor anything touching harness facts, `.agents\u002Fskills\u002Fharness-capture\u002FSKILL.md`\nis the procedure; `AGENTS.md` carries the hard rules.\n\n## Default posture\n\nAssume good faith and real frustration. Reframe a vague report into an\nanswerable technical question rather than bouncing it. Gather evidence before\nconcluding — including against your own first instinct, which will usually be\nthat the reporter is confused. Sometimes they are right and the adapter is\nwrong; that is the single most valuable thread type this project gets.\n\nResist scope expansion by momentum. \"It is only a small change\" is how a\nsingle-maintainer project acquires a permanent obligation.\n\n## Evidence standard\n\nUse the ladder the codebase already types (`ValidationMethod` in\n`packages\u002Fcore\u002Fsrc\u002Fadapter.ts`), strongest first: `captured`, `live-probe`,\n`router-log`, `schema-derived`, `type-derived`, `doc-derived`.\n\n- **Strong**: a hook-boundary payload, a fixture, a playback run, an observed\n  effect (a marker file written, a spawned command line read from a debug log).\n- **Weak**: a changelog, a release note, a vendor blog post, a type definition\n  read without running anything, \"it works for me\", a screenshot of a UI.\n\nTwo rules that have each already cost this project a wrong claim:\n\n1. **Router logs are not hook captures.** A harness may route a call under one\n   tool name and argument shape, then hand the hook a translated payload with a\n   different name, a different key, and fields dropped. Evidence about the\n   router is not evidence about the hook boundary. When a reporter conflates\n   them, say which one they have — without implying they wasted their time,\n   because router evidence is genuinely useful for defensive coverage.\n2. **\"The harness did not complain\" is not evidence.** Verify an effect by its\n   effect. Some harnesses print a completion line for hooks they skipped.\n\n## Decision matrix\n\n| Outcome | When |\n| --- | --- |\n| **Fix** | A defect on a supported surface with a reproduction, or a claim contradicted by evidence at or above the level the claim itself was recorded at. Minimal fix, no scope expansion. |\n| **Ask** | Exactly one missing fact would decide ownership. Ask for that one thing, not a form. |\n| **Route upstream** | The limitation is the harness's own. This is provable here — see below. |\n| **Decline** | Out of scope by `CONTRIBUTING.md`, or in scope but unaffordable. Say which. |\n\n### Route upstream, with proof\n\nThis project's strongest move, and it is not an assertion:\n\n```\nhooknostic inspect \u003Ctarget> --capability \u003Cid>\n```\n\nThat prints the adapter's own version-scoped rationale. Quote it, and give the\nreporter the command so they can re-run it themselves. Then offer to help word\nthe report for the harness's tracker — the reporter has hit a real wall, and\nrouting without that offer reads as a brush-off.\n\n### Ask for the smallest thing\n\nOnly what changes the ownership decision: `hooknostic doctor` output, the\nconfig, `check`\u002F`build` output with the HNxxx codes intact, or one capture. Do\nnot front-load a questionnaire onto someone who already filled in a form.\n\n### Decline cleanly\n\nCommon declines, each with a reason that is about the project rather than the\nperson:\n\n- Rating a capability higher than the adapter can deliver, because it \"mostly\n  works\". The rating *is* the product.\n- An emulation that hides a real difference. Emulation is welcome when declared\n  at the level it actually achieves, with a rationale saying how it differs.\n- A harness conditional in `core` or `runtime`. Harness knowledge lives in an\n  adapter.\n- Widening a `recommendedRange` on the strength of \"probably still works\".\n  Ranges are present-tense claims; extend them with a capture.\n- A shape table entry sourced from a changelog rather than an observation.\n- A new adapter with nobody behind it for the long run. Point at the\n  maintenance contract in `CONTRIBUTING.md` and ask directly.\n\n## Reviewing a pull request\n\nBeyond scope and correctness:\n\n- Was the new test **shown to fail** against the defect? Ask for that output if\n  it is missing. It is in the PR template for a reason.\n- Was the playback lane run for an adapter change? It costs the contributor\n  nothing, so there is no reason to accept its absence.\n- Committed example artifacts rebuilt from the repo root; `harness-support.md`\n  regenerated rather than edited; nothing generated hand-edited.\n- Provenance recorded honestly — a `validatedOn` record whose `method` matches\n  what was actually done.\n- A behavioural fix bundled with a broad refactor: ask to split, and say the\n  fix is welcome on its own.\n\n## Response style\n\nThankful, direct, factual. Name the boundary, cite the file or the command, give\nthe re-entry condition. No sarcasm, no blame, no speculative roadmap promises,\nand no implication that a question was stupid — the capability model is\ngenuinely subtle, and confusion about it is usually the docs' fault.\n\nClosing a thread: thank them, say what you checked, say why it is out of scope\nor upstream, name the supported surface, and state what would reopen it.\n\n## Re-entry criteria\n\nReopen when: a reproduction lands on a supported surface; a capture arrives that\nsettles a previously unevidenced claim; a PR narrows to supported behaviour with\ntests; or a new harness release changes the behaviour the decision rested on —\nthat last one is expected, and a closed thread is not a permanent verdict about\nthe harness, only about the versions validated at the time.\n\n## Anti-patterns\n\n- Accepting a plausible claim because the reporter sounded confident.\n- Letting a low-fidelity thread sit open indefinitely instead of asking once and\n  closing kindly.\n- Treating a vendor changelog as capture-grade evidence.\n- Filing router evidence as a hook capture, or vice versa.\n- Agreeing to an adapter without asking who maintains it.\n- Fixing a reported symptom in `core` because that is where the change is\n  smallest.\n- Restating the scope statement in a reply instead of linking it, so that the\n  two drift apart.\n","# Community triage\n\nYou are triaging for a **single-maintainer** project whose product is honest\ncapability information, not maximal coverage. Most threads are decided by scope\nand evidence rather than by effort, and saying so kindly and early is the\nservice — a request left open for months is worse for the reporter than a clear\n\"not this project, and here is why\".\n\n## Required context\n\nThe canonical scope statement lives in `CONTRIBUTING.md` (\"What Hooknostic is\" \u002F\n\"What Hooknostic is not\" \u002F \"The rule that decides most contributions\"). Read it\nand **link to it**; do not restate it in your reply, and never let this file\ndrift into a second copy of it.\n\nFor anything touching harness facts, `.agents\u002Fskills\u002Fharness-capture\u002FSKILL.md`\nis the procedure; `AGENTS.md` carries the hard rules.\n\n## Default posture\n\nAssume good faith and real frustration. Reframe a vague report into an\nanswerable technical question rather than bouncing it. Gather evidence before\nconcluding — including against your own first instinct, which will usually be\nthat the reporter is confused. Sometimes they are right and the adapter is\nwrong; that is the single most valuable thread type this project gets.\n\nResist scope expansion by momentum. \"It is only a small change\" is how a\nsingle-maintainer project acquires a permanent obligation.\n\n## Evidence standard\n\nUse the ladder the codebase already types (`ValidationMethod` in\n`packages\u002Fcore\u002Fsrc\u002Fadapter.ts`), strongest first: `captured`, `live-probe`,\n`router-log`, `schema-derived`, `type-derived`, `doc-derived`.\n\n- **Strong**: a hook-boundary payload, a fixture, a playback run, an observed\n  effect (a marker file written, a spawned command line read from a debug log).\n- **Weak**: a changelog, a release note, a vendor blog post, a type definition\n  read without running anything, \"it works for me\", a screenshot of a UI.\n\nTwo rules that have each already cost this project a wrong claim:\n\n1. **Router logs are not hook captures.** A harness may route a call under one\n   tool name and argument shape, then hand the hook a translated payload with a\n   different name, a different key, and fields dropped. Evidence about the\n   router is not evidence about the hook boundary. When a reporter conflates\n   them, say which one they have — without implying they wasted their time,\n   because router evidence is genuinely useful for defensive coverage.\n2. **\"The harness did not complain\" is not evidence.** Verify an effect by its\n   effect. Some harnesses print a completion line for hooks they skipped.\n\n## Decision matrix\n\n| Outcome | When |\n| --- | --- |\n| **Fix** | A defect on a supported surface with a reproduction, or a claim contradicted by evidence at or above the level the claim itself was recorded at. Minimal fix, no scope expansion. |\n| **Ask** | Exactly one missing fact would decide ownership. Ask for that one thing, not a form. |\n| **Route upstream** | The limitation is the harness's own. This is provable here — see below. |\n| **Decline** | Out of scope by `CONTRIBUTING.md`, or in scope but unaffordable. Say which. |\n\n### Route upstream, with proof\n\nThis project's strongest move, and it is not an assertion:\n\n```\nhooknostic inspect \u003Ctarget> --capability \u003Cid>\n```\n\nThat prints the adapter's own version-scoped rationale. Quote it, and give the\nreporter the command so they can re-run it themselves. Then offer to help word\nthe report for the harness's tracker — the reporter has hit a real wall, and\nrouting without that offer reads as a brush-off.\n\n### Ask for the smallest thing\n\nOnly what changes the ownership decision: `hooknostic doctor` output, the\nconfig, `check`\u002F`build` output with the HNxxx codes intact, or one capture. Do\nnot front-load a questionnaire onto someone who already filled in a form.\n\n### Decline cleanly\n\nCommon declines, each with a reason that is about the project rather than the\nperson:\n\n- Rating a capability higher than the adapter can deliver, because it \"mostly\n  works\". The rating *is* the product.\n- An emulation that hides a real difference. Emulation is welcome when declared\n  at the level it actually achieves, with a rationale saying how it differs.\n- A harness conditional in `core` or `runtime`. Harness knowledge lives in an\n  adapter.\n- Widening a `recommendedRange` on the strength of \"probably still works\".\n  Ranges are present-tense claims; extend them with a capture.\n- A shape table entry sourced from a changelog rather than an observation.\n- A new adapter with nobody behind it for the long run. Point at the\n  maintenance contract in `CONTRIBUTING.md` and ask directly.\n\n## Reviewing a pull request\n\nBeyond scope and correctness:\n\n- Was the new test **shown to fail** against the defect? Ask for that output if\n  it is missing. It is in the PR template for a reason.\n- Was the playback lane run for an adapter change? It costs the contributor\n  nothing, so there is no reason to accept its absence.\n- Committed example artifacts rebuilt from the repo root; `harness-support.md`\n  regenerated rather than edited; nothing generated hand-edited.\n- Provenance recorded honestly — a `validatedOn` record whose `method` matches\n  what was actually done.\n- A behavioural fix bundled with a broad refactor: ask to split, and say the\n  fix is welcome on its own.\n\n## Response style\n\nThankful, direct, factual. Name the boundary, cite the file or the command, give\nthe re-entry condition. No sarcasm, no blame, no speculative roadmap promises,\nand no implication that a question was stupid — the capability model is\ngenuinely subtle, and confusion about it is usually the docs' fault.\n\nClosing a thread: thank them, say what you checked, say why it is out of scope\nor upstream, name the supported surface, and state what would reopen it.\n\n## Re-entry criteria\n\nReopen when: a reproduction lands on a supported surface; a capture arrives that\nsettles a previously unevidenced claim; a PR narrows to supported behaviour with\ntests; or a new harness release changes the behaviour the decision rested on —\nthat last one is expected, and a closed thread is not a permanent verdict about\nthe harness, only about the versions validated at the time.\n\n## Anti-patterns\n\n- Accepting a plausible claim because the reporter sounded confident.\n- Letting a low-fidelity thread sit open indefinitely instead of asking once and\n  closing kindly.\n- Treating a vendor changelog as capture-grade evidence.\n- Filing router evidence as a hook capture, or vice versa.\n- Agreeing to an adapter without asking who maintains it.\n- Fixing a reported symptom in `core` because that is where the change is\n  smallest.\n- Restating the scope statement in a reply instead of linking it, so that the\n  two drift apart.\n",{"name":5,"description":7},"master","b8669236433862d74703a9a146bf1373f9b0b209",".agents\u002Fskills\u002Fcommunity-triage",0,"https:\u002F\u002Fgithub.com\u002Froryeckel\u002Fhooknostic","https:\u002F\u002Fgithub.com\u002Froryeckel","https:\u002F\u002F2labz.com","I write a lot of software and delegate the rest to LLMs","2026-11-15 04:44:54",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":45},[35,38,41],{"tool":36,"command":37},"npx","npx skills add roryeckel\u002Fhooknostic --skill community-triage",{"tool":39,"command":40},"bunx","bunx skills add roryeckel\u002Fhooknostic --skill community-triage",{"tool":42,"command":43},"pnpm","pnpm dlx skills add roryeckel\u002Fhooknostic --skill community-triage","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Froryeckel\u002Fhooknostic\u002Fblob\u002Fmaster\u002F.agents\u002Fskills\u002Fcommunity-triage\u002FSKILL.md",true,[],[48,55,60,65,70,77],{"skillNo":49,"name":50,"nameZh":6,"description":51,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":52,"categories":53,"keywords":54,"publishedAt":19},"01M7AXANW27HCE8N4ZD2H8GZG5","sample","Explain the synthetic sample project.","https:\u002F\u002Fgithub.com\u002Froryeckel\u002Fhooknostic\u002Fblob\u002Fmaster\u002Fexamples\u002Flocal-project\u002Fskills\u002Fsample\u002FSKILL.md",[],[],{"skillNo":56,"name":50,"nameZh":6,"description":51,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":57,"categories":58,"keywords":59,"publishedAt":19},"01M7AXAKKRACEVMN7CZ44ZC6A1","https:\u002F\u002Fgithub.com\u002Froryeckel\u002Fhooknostic\u002Fblob\u002Fmaster\u002Fexamples\u002Flocal-project\u002F.pi\u002Fskills\u002Fsample\u002FSKILL.md",[],[],{"skillNo":61,"name":50,"nameZh":6,"description":51,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":62,"categories":63,"keywords":64,"publishedAt":19},"01M7AXAJEYX4DKRCRBSNPM7J09","https:\u002F\u002Fgithub.com\u002Froryeckel\u002Fhooknostic\u002Fblob\u002Fmaster\u002Fexamples\u002Flocal-project\u002F.claude\u002Fskills\u002Fsample\u002FSKILL.md",[],[],{"skillNo":66,"name":50,"nameZh":6,"description":51,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":67,"categories":68,"keywords":69,"publishedAt":19},"01M7AXAFPMXR8EJM4D2Z1WKBH2","https:\u002F\u002Fgithub.com\u002Froryeckel\u002Fhooknostic\u002Fblob\u002Fmaster\u002Fexamples\u002Flocal-project\u002F.agents\u002Fskills\u002Fsample\u002FSKILL.md",[],[],{"skillNo":71,"name":72,"nameZh":6,"description":73,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":74,"categories":75,"keywords":76,"publishedAt":19},"01M7AXADDP9ASKQBVG8T8KE3XV","greet","Greet the user by name and summarize the repo state.","https:\u002F\u002Fgithub.com\u002Froryeckel\u002Fhooknostic\u002Fblob\u002Fmaster\u002Fexamples\u002Fagent-plugin\u002Fskills\u002Fgreet\u002FSKILL.md",[],[],{"skillNo":78,"name":79,"nameZh":6,"description":80,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":81,"categories":82,"keywords":83,"publishedAt":19},"01M7AXAC2V6M2XGEEBAPARQA4R","where","Reports where this skill lives. Use when asked where the probe skill is.","https:\u002F\u002Fgithub.com\u002Froryeckel\u002Fhooknostic\u002Fblob\u002Fmaster\u002F.capture\u002Fskill-directory\u002FSKILL.md",[],[]]