[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6KWSGTCNCCSWBEKRHPG10DV":3,"skill-same:01M6KWSGTCNCCSWBEKRHPG10DV":63},{"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":29,"linkOnly":30,"markdownContent":31,"markdownBody":32,"frontMatter":33,"refName":35,"commitSha":36,"skillDirectory":37,"forksCount":38,"repositoryHtmlUrl":39,"repositoryHomepageUrl":6,"ownerHtmlUrl":40,"ownerBlogUrl":41,"ownerDescription":42,"updatedAt":43,"install":44,"resources":56},"01M6KWSGTCNCCSWBEKRHPG10DV","ask-harness",null,"Find which skill or flow fits your situation, with a map of every harness-kit skill.","dougborg\u002Fharness-kit","dougborg","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F1261222?v=4","01M6KWS4PB51EYD4ZDJJSGNGB2","Doug Borg","01M6KWS4T8ZZW6SX5QZS9B5BN2",0,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fdougborg\u002Fharness-kit\u002Fblob\u002Fmain\u002Fclaude-skills\u002Fask-harness\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28],"harness","ask","ask harness","flow","find","map","with","kit","true","which","2026-11-10 10:12:30",false,"---\nname: ask-harness\ndescription: Find which skill or flow fits your situation, with a map of every harness-kit skill.\ndisable-model-invocation: true\n---\n\n# Ask Harness\n\nYou don't need to remember every skill. This is the map: one **main flow**\nmost work travels, the **on-ramps** that feed it, the lifecycle around pull\nrequests and issues, and the standalone tools. A name with a slash is one you\ntype to start (`\u002Fname` on Claude Code, `$name` on Codex); a bare name is one\nthe agent also picks up on its own when the task fits, and you can still ask\nfor it by name.\n\n## The main flow: idea to merged change\n\n1. **Align.** `\u002Fgrill-with-docs` interviews you until every branch of the\n   design is settled, and records the project's terms in `GLOSSARY.md` and\n   hard-to-reverse decisions as ADRs. Use `\u002Fgrill-me` when there's no repo to\n   record into. Both run the `grilling` skill underneath.\n2. **Settle what talk can't.** When a question needs a runnable answer, build a\n   cheap throwaway artifact to react to with `prototype` (a state-model demo\n   or UI variants; in a bigger effort, a wayfinder prototype ticket); when it\n   needs outside facts, the `research` skill reads\n   primary sources and cites them.\n3. **Specify.** For work that spans sessions, `\u002Fto-spec` turns the\n   conversation into a spec issue, and `\u002Fto-tickets` splits it into\n   tracer-bullet sub-issues with blocking links. Small work skips straight to\n   step 4.\n4. **Build.** `\u002Fimplement` takes one ticket test-first (`tdd`, `minimal-change`)\n   to an open PR. `\u002Fimplement-spec` builds a whole spec's tickets in parallel\n   worktrees onto one integration branch.\n5. **Ship.** `open-pr` simplifies the diff, writes the body (`pr-body`), runs\n   the standards and spec review passes (`review-pr`, `code-reviewer`), and\n   waits for CI.\n6. **Improve the environment.** `\u002Fharness retro` looks back over the session\n   for missing checks, standards, pointers, and access, and sends each\n   lesson to the project's facts, a project-local skill, or upstream, never\n   into a vendored file. `\u002Fsession-retro`\n   records what the work itself taught. Run them in the session they look\n   back on, before you clear it.\n\nKeep steps 1 to 3 in one context window so the spec builds on the actual\nreasoning; each `\u002Fimplement` can then start fresh from its ticket. If the\nwindow fills before `\u002Fto-tickets` (reasoning gets noticeably weaker past\nroughly 150k tokens on current models), compact at the nearest phase boundary\nrather than pushing on.\n\n## On-ramps\n\n- **Issues and PRs piling up from others**: `\u002Ftriage` moves them through\n  triage states to agent-ready briefs, which `\u002Fimplement` later picks up.\n  Tickets from `\u002Fto-tickets` are ready already.\n- **Something's broken**: `diagnosing-bugs` builds a fast check that goes red\n  on the bug before any theory, then fixes the root cause with a regression\n  test.\n- **The code has got hard to change**: `\u002Fimprove-codebase-architecture`\n  finds shallow modules worth deepening, shows them in a visual report, and\n  grills through the one you pick.\n- **Too big and foggy for one session**: `\u002Fwayfinder` charts a map of decision\n  tickets and settles them one per session; when the way is clear, it hands\n  off to `\u002Fto-spec`.\n\n## Around pull requests and issues\n\n| You want to | Use |\n| --- | --- |\n| Commit with the project's checks | `commit` |\n| Open a PR and see it through review and CI | `open-pr` |\n| Work through or consolidate several dependency PRs | `open-pr` |\n| Review a PR, or answer its review comments | `review-pr`; `\u002Fpr-comments` for a reply-only pass |\n| Bring a branch up to date | `\u002Frebase` |\n| File, update, or close an issue | `issue-create`, `issue-update`, `issue-close` |\n| Split or merge issues | `\u002Fissue-restructure` |\n| Decide what to work on next | `\u002Fgroom` (it also surveys `shortcut-ledger`) |\n| Recap your day | `\u002Fstandup` |\n| Coordinate several agents or people | `agent-standup`; on Claude Code, `budget` before a big fan-out |\n\n## Vocabulary underneath\n\n- `domain-modeling`: sharpen the project's terms and record them.\n- `codebase-design`: the deep-module vocabulary (module, interface, depth,\n  seam) for shaping a module.\n\n## Standalone\n\n- `\u002Fto-questionnaire`: when the answer is in someone else's head, write them a\n  questionnaire.\n- `\u002Fwait-what`: the last message didn't land; re-explain it plainly.\n- `\u002Fadhd-mode`: action-first responses for the rest of the session (on\n  Claude Code, the ADHD output style does it without invoking anything).\n- `\u002Fhandoff`: move the work to another host, directory, or person.\n- `\u002Fteach`: learn a topic over several sessions, with lessons and a record\n  of what you've learned.\n- `\u002Floop-me` (experimental): find recurring work worth delegating and spec\n  it as workflows.\n- `ui-review`: accessibility and UX audit of web UI.\n- `wizard`: a script that walks a person through steps only they can do,\n  such as creating accounts and copying API keys into `.env` and CI secrets.\n- `documentation-writer` and `\u002Fsvg-logo-designer`: docs and logos.\n- `skill-writer`, `harness`, `\u002Fharness-builder`, `harness-issue`: maintain\n  the harness itself. The `harness` skill audits, bootstraps, updates, and\n  improves it, and migrates a copy-and-edit harness to a vendored snapshot;\n  once vendored, updates arrive as pull requests from the project's\n  scheduled update workflow.\n\n## Between phases\n\nAt the boundary between two chunks of work (grilling done, implementation\ndone), choose how to carry on: continue, clear, hand off, send it to a\nsubagent, or compact. [PHASE-BOUNDARIES.md](PHASE-BOUNDARIES.md) has the\norder to ask in and why. Mid-phase, keep going or split the rest into\nsubagents.\n","# Ask Harness\n\nYou don't need to remember every skill. This is the map: one **main flow**\nmost work travels, the **on-ramps** that feed it, the lifecycle around pull\nrequests and issues, and the standalone tools. A name with a slash is one you\ntype to start (`\u002Fname` on Claude Code, `$name` on Codex); a bare name is one\nthe agent also picks up on its own when the task fits, and you can still ask\nfor it by name.\n\n## The main flow: idea to merged change\n\n1. **Align.** `\u002Fgrill-with-docs` interviews you until every branch of the\n   design is settled, and records the project's terms in `GLOSSARY.md` and\n   hard-to-reverse decisions as ADRs. Use `\u002Fgrill-me` when there's no repo to\n   record into. Both run the `grilling` skill underneath.\n2. **Settle what talk can't.** When a question needs a runnable answer, build a\n   cheap throwaway artifact to react to with `prototype` (a state-model demo\n   or UI variants; in a bigger effort, a wayfinder prototype ticket); when it\n   needs outside facts, the `research` skill reads\n   primary sources and cites them.\n3. **Specify.** For work that spans sessions, `\u002Fto-spec` turns the\n   conversation into a spec issue, and `\u002Fto-tickets` splits it into\n   tracer-bullet sub-issues with blocking links. Small work skips straight to\n   step 4.\n4. **Build.** `\u002Fimplement` takes one ticket test-first (`tdd`, `minimal-change`)\n   to an open PR. `\u002Fimplement-spec` builds a whole spec's tickets in parallel\n   worktrees onto one integration branch.\n5. **Ship.** `open-pr` simplifies the diff, writes the body (`pr-body`), runs\n   the standards and spec review passes (`review-pr`, `code-reviewer`), and\n   waits for CI.\n6. **Improve the environment.** `\u002Fharness retro` looks back over the session\n   for missing checks, standards, pointers, and access, and sends each\n   lesson to the project's facts, a project-local skill, or upstream, never\n   into a vendored file. `\u002Fsession-retro`\n   records what the work itself taught. Run them in the session they look\n   back on, before you clear it.\n\nKeep steps 1 to 3 in one context window so the spec builds on the actual\nreasoning; each `\u002Fimplement` can then start fresh from its ticket. If the\nwindow fills before `\u002Fto-tickets` (reasoning gets noticeably weaker past\nroughly 150k tokens on current models), compact at the nearest phase boundary\nrather than pushing on.\n\n## On-ramps\n\n- **Issues and PRs piling up from others**: `\u002Ftriage` moves them through\n  triage states to agent-ready briefs, which `\u002Fimplement` later picks up.\n  Tickets from `\u002Fto-tickets` are ready already.\n- **Something's broken**: `diagnosing-bugs` builds a fast check that goes red\n  on the bug before any theory, then fixes the root cause with a regression\n  test.\n- **The code has got hard to change**: `\u002Fimprove-codebase-architecture`\n  finds shallow modules worth deepening, shows them in a visual report, and\n  grills through the one you pick.\n- **Too big and foggy for one session**: `\u002Fwayfinder` charts a map of decision\n  tickets and settles them one per session; when the way is clear, it hands\n  off to `\u002Fto-spec`.\n\n## Around pull requests and issues\n\n| You want to | Use |\n| --- | --- |\n| Commit with the project's checks | `commit` |\n| Open a PR and see it through review and CI | `open-pr` |\n| Work through or consolidate several dependency PRs | `open-pr` |\n| Review a PR, or answer its review comments | `review-pr`; `\u002Fpr-comments` for a reply-only pass |\n| Bring a branch up to date | `\u002Frebase` |\n| File, update, or close an issue | `issue-create`, `issue-update`, `issue-close` |\n| Split or merge issues | `\u002Fissue-restructure` |\n| Decide what to work on next | `\u002Fgroom` (it also surveys `shortcut-ledger`) |\n| Recap your day | `\u002Fstandup` |\n| Coordinate several agents or people | `agent-standup`; on Claude Code, `budget` before a big fan-out |\n\n## Vocabulary underneath\n\n- `domain-modeling`: sharpen the project's terms and record them.\n- `codebase-design`: the deep-module vocabulary (module, interface, depth,\n  seam) for shaping a module.\n\n## Standalone\n\n- `\u002Fto-questionnaire`: when the answer is in someone else's head, write them a\n  questionnaire.\n- `\u002Fwait-what`: the last message didn't land; re-explain it plainly.\n- `\u002Fadhd-mode`: action-first responses for the rest of the session (on\n  Claude Code, the ADHD output style does it without invoking anything).\n- `\u002Fhandoff`: move the work to another host, directory, or person.\n- `\u002Fteach`: learn a topic over several sessions, with lessons and a record\n  of what you've learned.\n- `\u002Floop-me` (experimental): find recurring work worth delegating and spec\n  it as workflows.\n- `ui-review`: accessibility and UX audit of web UI.\n- `wizard`: a script that walks a person through steps only they can do,\n  such as creating accounts and copying API keys into `.env` and CI secrets.\n- `documentation-writer` and `\u002Fsvg-logo-designer`: docs and logos.\n- `skill-writer`, `harness`, `\u002Fharness-builder`, `harness-issue`: maintain\n  the harness itself. The `harness` skill audits, bootstraps, updates, and\n  improves it, and migrates a copy-and-edit harness to a vendored snapshot;\n  once vendored, updates arrive as pull requests from the project's\n  scheduled update workflow.\n\n## Between phases\n\nAt the boundary between two chunks of work (grilling done, implementation\ndone), choose how to carry on: continue, clear, hand off, send it to a\nsubagent, or compact. [PHASE-BOUNDARIES.md](PHASE-BOUNDARIES.md) has the\norder to ask in and why. Mid-phase, keep going or split the rest into\nsubagents.\n",{"name":5,"description":7,"disable-model-invocation":34},true,"main","023b01bfb4a86fbf1e38a749f1044cce3e5da67d","claude-skills\u002Fask-harness",1,"https:\u002F\u002Fgithub.com\u002Fdougborg\u002Fharness-kit","https:\u002F\u002Fgithub.com\u002Fdougborg","http:\u002F\u002Fdougborg.org","Platforms and Infrastructure Software Engineer","2026-11-13 07:43:37",{"source":8,"skillName":5,"commands":45,"prompt":55,"zipAvailable":34},[46,49,52],{"tool":47,"command":48},"npx","npx skills add dougborg\u002Fharness-kit --skill ask-harness",{"tool":50,"command":51},"bunx","bunx skills add dougborg\u002Fharness-kit --skill ask-harness",{"tool":53,"command":54},"pnpm","pnpm dlx skills add dougborg\u002Fharness-kit --skill ask-harness","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fdougborg\u002Fharness-kit\u002Fblob\u002Fmain\u002Fclaude-skills\u002Fask-harness\u002FSKILL.md",[57],{"path":58,"resourceType":59,"resourceTypeDesc":60,"sizeBytes":61,"sourceHtmlUrl":62},"PHASE-BOUNDARIES.md","REFERENCE","参考文档",2308,"https:\u002F\u002Fgithub.com\u002Fdougborg\u002Fharness-kit\u002Fblob\u002Fmain\u002Fclaude-skills\u002Fask-harness\u002FPHASE-BOUNDARIES.md",[64,82,103,136,158,175],{"skillNo":65,"name":66,"nameZh":6,"description":67,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":68,"categories":69,"keywords":75,"publishedAt":29},"01M6KX4J65S5FZKMSN5DQNPSZM","teach","Teach yourself a topic over several sessions, with lessons, references, and a record of what you've learned.","https:\u002F\u002Fgithub.com\u002Fdougborg\u002Fharness-kit\u002Fblob\u002Fmain\u002Fskills\u002Fwriting\u002Fteach\u002FSKILL.md",[70],{"categoryNo":71,"categoryName":72,"taxonomyType":73,"taxonomyTypeDesc":74},"01JBSEEDCAT000000000000039","教师与教育","PROFESSION","适用职业",[66,76,77,25,78,79,80,81],"lessons","you","learned","record","references","sessions",{"skillNo":83,"name":84,"nameZh":6,"description":85,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":93,"publishedAt":29},"01M6KX4FSSG7MHF00JDT5QT0BQ","svg-logo-designer","Generate SVG logos with multiple concepts, layouts, and color variations.","https:\u002F\u002Fgithub.com\u002Fdougborg\u002Fharness-kit\u002Fblob\u002Fmain\u002Fskills\u002Fwriting\u002Fsvg-logo-designer\u002FSKILL.md",[88],{"categoryNo":89,"categoryName":90,"taxonomyType":91,"taxonomyTypeDesc":92},"01JBSEEDCAT000000000000016","图像与多媒体","SCENARIO","使用场景",[94,95,96,97,98,99,25,100,101,102],"logo","designer","svg logo designer","SVG","color","concepts","write","read","generate",{"skillNo":104,"name":105,"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":115,"publishedAt":29},"01M6KX4BD3J9MZPPSH4MENH7FP","documentation-writer","Writes and revises human-facing documentation — READMEs, contributor and setup guides, reference pages, architecture notes, CLAUDE.md, and changelog entries. Covers picking the right document type for the question being asked, structuring for skimming, writing examples that actually run, and keeping docs from going stale. Use when creating or rewriting docs, when a README has drifted from the code, or when deciding where a piece of information belongs. For SKILL.md and agent files, use \u002Fskill-writer instead.","https:\u002F\u002Fgithub.com\u002Fdougborg\u002Fharness-kit\u002Fblob\u002Fmain\u002Fskills\u002Fwriting\u002Fdocumentation-writer\u002FSKILL.md",[109,112],{"categoryNo":110,"categoryName":111,"taxonomyType":91,"taxonomyTypeDesc":92},"01JBSEEDCAT000000000000018","写作与文案",{"categoryNo":113,"categoryName":114,"taxonomyType":73,"taxonomyTypeDesc":74},"01JBSEEDCAT000000000000036","内容创作者",[116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135],"writer","documentation writer","documentation","entries","changelog","architecture","document","files","use","question","README","notes","code","examples","pages","for","information","guides","type","writing",{"skillNo":137,"name":138,"nameZh":6,"description":139,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":140,"categories":141,"keywords":142,"publishedAt":29},"01M6KX47ZJ4ZWHDN3MY8SYAWZZ","adhd-mode","Shapes every response for a reader with ADHD: the next action first, numbered steps, progress restated each turn, tangents held back, concrete time estimates, wins made visible. Stays on for the rest of the session until the reader says \"stop adhd mode\". On Claude Code, the harness-kit:adhd output style does the same without invoking this skill.","https:\u002F\u002Fgithub.com\u002Fdougborg\u002Fharness-kit\u002Fblob\u002Fmain\u002Fskills\u002Fwriting\u002Fadhd-mode\u002FSKILL.md",[],[143,144,145,146,131,147,26,25,148,149,150,151,152,153,154,155,156,157,19,128],"adhd","mode","adhd mode","next","action","output","style","progress","time","back","stop","session","reader","first","REST",{"skillNo":159,"name":160,"nameZh":6,"description":161,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":162,"categories":163,"keywords":164,"publishedAt":29},"01M6KX463Z6E18EPDW873GDHKB","wait-what","That last message didn't land: re-explain it with the missing context.","https:\u002F\u002Fgithub.com\u002Fdougborg\u002Fharness-kit\u002Fblob\u002Fmain\u002Fskills\u002Fthinking\u002Fwait-what\u002FSKILL.md",[],[165,166,167,25,168,169,170,171,172,173,174],"wait","wait what","re","it","missing","message","context","explain","land","last",{"skillNo":176,"name":177,"nameZh":6,"description":178,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":179,"categories":180,"keywords":181,"publishedAt":29},"01M6KX43NE1NB0Y2HECY515730","to-questionnaire","Turn a decision you can't make alone into a questionnaire for the one person who can answer it, filled in async or together in a meeting.","https:\u002F\u002Fgithub.com\u002Fdougborg\u002Fharness-kit\u002Fblob\u002Fmain\u002Fskills\u002Fthinking\u002Fto-questionnaire\u002FSKILL.md",[],[182,183,131,184,185,186,187,77,188,168,189,190],"to questionnaire","questionnaire","person","can","answer","meeting","make","async","decision"]