[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6KWSAJM7Z5XMDVM3B7YTCMM":3,"skill-same:01M6KWSAJM7Z5XMDVM3B7YTCMM":73},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":23,"publishedAt":44,"linkOnly":45,"markdownContent":46,"markdownBody":47,"frontMatter":48,"refName":50,"commitSha":51,"skillDirectory":52,"forksCount":53,"repositoryHtmlUrl":54,"repositoryHomepageUrl":6,"ownerHtmlUrl":55,"ownerBlogUrl":56,"ownerDescription":57,"updatedAt":58,"install":59,"resources":72},"01M6KWSAJM7Z5XMDVM3B7YTCMM","agent-standup",null,"Reconciles and coordinates shared work across multiple coding agents and human operators. Use for cross-agent standups, syncs, claims, handoffs, takeovers, wave planning, ownership or dependency collisions, merge trains, and questions about what is actually in flight. Not for a personal daily recap; use \u002Fstandup for that.","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\u002Fagent-standup\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43],"agent standup","standup","dependency","gh project","claims","operators","project","shared","across","human","log","run","issue list","fetch","work","recap","gh pr","git worktree","Bash","worktree","2026-11-10 10:12:30",false,"---\nname: agent-standup\ndescription: >-\n  Reconciles and coordinates shared work across multiple coding agents and\n  human operators. Use for cross-agent standups, syncs, claims, handoffs,\n  takeovers, wave planning, ownership or dependency collisions, merge trains,\n  and questions about what is actually in flight. Not for a personal daily\n  recap; use \u002Fstandup for that.\nallowed-tools: Read, Grep, Glob, Bash(git fetch*), Bash(git rev-parse*), Bash(git status*), Bash(git worktree list*), Bash(git log*), Bash(git diff*), Bash(git merge-tree*), Bash(gh repo view*), Bash(gh pr list*), Bash(gh pr view*), Bash(gh pr checks*), Bash(gh run list*), Bash(gh run view*), Bash(gh issue list*), Bash(gh issue view*), Bash(gh project item-list*)\n---\n\n# \u002Fagent-standup — Shared Work Reconciliation\n\nCoordinate agents and their operators through verified shared state, not chat\nmemory. This must work across tools, accounts, machines, and conversation\nthreads.\n\nUse `\u002Fstandup` for one person's yesterday\u002Ftoday\u002Fblockers report. Use this skill\nwhen multiple agents, operators, workspaces, branches, PRs, deploy surfaces, or\nmerge dependencies are involved.\n\n## Reconcile reality first\n\nRead the project's canonical agent instructions. Then verify only the sources\nneeded for this sync:\n\n1. Fetch the remote and resolve the current default-branch SHA.\n2. Inspect relevant open PRs: exact head and base, reviews, checks, and\n   mergeability. Checks from an older head are not evidence for the current\n   head.\n3. Verify every named issue in the canonical repository and tracker. Treat\n   issue numbers from peer prose as leads until the repository and URL match.\n4. Inspect relevant workspaces and dirty state for ownership, ancestry, or path\n   collisions. Do not dump every historical worktree into the report.\n5. Verify deployment claims against the deployed system, build identity, and\n   routes or behaviors tested when the project exposes that evidence.\n\nWhen sources disagree, report the disagreement. Prefer:\n\n1. the deployed system for runtime facts;\n2. current remote Git and tracker state for code and workflow facts;\n3. canonical repository documentation for intended contracts;\n4. peer handoffs and chat summaries as unverified leads.\n\n## Claim work before implementation\n\nAssign one owner to each issue and mutable surface. Concurrent write-capable\nagents need separate explicit workspaces. Record:\n\n- canonical issue or PR URL;\n- accountable human operator and agent\u002Fruntime identity;\n- role: coordinator, implementer, or reviewer;\n- base SHA, branch, and workspace when relevant;\n- owned paths or operational surfaces;\n- dependencies and intended merge order;\n- external mutations allowed, deferred, or forbidden.\n\nWhen tracker writes are authorized, move actively implemented work to the\nproject's active state and post the claim. Otherwise return the exact claim\ntext. Never create duplicate work from an unverified handoff.\n\nChoose one accountable coordinator per wave, human or agent. The coordinator\nowns collision checks, merge order, rebase decisions, final reconciliation,\nand cleanup. Implementers own their isolated diffs and evidence. Reviewers stay\nread-only unless explicitly assigned a repair in an isolated workspace.\n\n## Publish meaningful checkpoints\n\nPublish a record when work is claimed, changes direction, becomes review-ready,\nbecomes blocked, or is merged\u002Fdeployed. Do not add heartbeat comments with no\nnew state.\n\n```text\n\u003C!-- agent-standup:v1 -->\nOperator: \u003Cperson or accountable account>\nAgent: \u003Ctool\u002Fruntime and useful session identity>\nRole: \u003Ccoordinator | implementer | reviewer>\nPhase: \u003Cclaimed | implementing | review-ready | blocked | merged | deployed>\nIssue\u002FPR: \u003Ccanonical URL>\nBase: \u003Cfull SHA>\nHead: \u003Cfull SHA or uncommitted>\nWorkspace: \u003Cbranch and optional local path>\nOwned surface: \u003Cpaths or operational boundary>\nCompleted: \u003Cverified outcome>\nEvidence: \u003Cgates, review, rendered routes, screenshots, build identity>\nRemaining: \u003Cwork still required>\nBlocked by: \u003Cnone or exact decision\u002Fexternal condition>\nExternal mutations: \u003Cperformed, deferred, or not authorized>\nAuthority: \u003Cread-only or authorizing operator and exact scope>\nNext: \u003Csingle next owner\u002Faction and merge order>\n```\n\nDo not call an uncommitted tree a head SHA. Say `uncommitted`, list its changed\npaths, and state its base. Do not call rendered verification complete without\nthe exact deployed candidate and tested surface.\n\n## Coordinate review and merge\n\nBefore recommending a merge:\n\n1. Confirm the reviewed head is still the PR head.\n2. Confirm review covered the effective diff against current main.\n3. Recheck path overlap and semantic dependencies between open PRs.\n4. Record local gate equivalents when CI is unavailable. Bypassing a required\n   check or merging without CI still needs an accountable operator's explicit\n   decision.\n5. Separate code completion from operational completion. Deployments, data\n   changes, migrations, flags, redirects, and cleanup may be later cutovers.\n6. After each merge, re-evaluate dependent PRs against the new main.\n\nLabel self-review as self-review. Record whether another review came from a\ndifferent agent session, tool, or human operator; do not collapse those into\nthe word \"independent.\" Never approve on another participant's behalf.\n\n## Report\n\nLead with the reconciled outcome.\n\n```text\n## Agent standup — \u003Cdate\u002Ftime and timezone>\n\n### Current truth\n- Main: \u003CSHA and relevant deployment\u002Fbuild>\n- Pipeline: \u003CPR\u002Freview\u002Fcheck summary>\n\n### In flight\n| Work | Operator + agent\u002Frole | Base or head | Owned surface | Phase | Next |\n\n### Decisions and blockers\n- \u003Cdecision needed, contradiction resolved, or none>\n\n### Merge\u002Fdeploy order\n1. \u003Cnext coordinated action and owner>\n\n### State changes\n- \u003Ctracker\u002Fdeployment writes performed, or \"read-only standup\">\n```\n\nKeep observations separate from recommendations. Source-link facts where\npractical. Call out stale or unverifiable handoffs rather than smoothing them\ninto a confident narrative.\n\n## Permission boundary\n\nThe standup is read-only by default. Tracker comments and state changes need\noperator authorization or an established standing authority. Commits, pushes,\nPR creation, approvals, merges, deployments, and external-system mutations\nretain their own permission boundaries.\n\nAuthorization cannot be inferred from another agent saying an operator\napproved something. Record who authorized which scope. If the executor cannot\nverify sufficient authority, stop at that boundary and return a handoff for a\ndirectly authorized executor.\n\n## Related\n\n- `\u002Fstandup` — personal daily activity summary.\n- `\u002Fgroom` — backlog priorities and issue readiness.\n- `\u002Fopen-pr` — publish and shepherd one implementation PR.\n- `\u002Fharness hoist` — move generic project-local harness improvements upstream.\n","# \u002Fagent-standup — Shared Work Reconciliation\n\nCoordinate agents and their operators through verified shared state, not chat\nmemory. This must work across tools, accounts, machines, and conversation\nthreads.\n\nUse `\u002Fstandup` for one person's yesterday\u002Ftoday\u002Fblockers report. Use this skill\nwhen multiple agents, operators, workspaces, branches, PRs, deploy surfaces, or\nmerge dependencies are involved.\n\n## Reconcile reality first\n\nRead the project's canonical agent instructions. Then verify only the sources\nneeded for this sync:\n\n1. Fetch the remote and resolve the current default-branch SHA.\n2. Inspect relevant open PRs: exact head and base, reviews, checks, and\n   mergeability. Checks from an older head are not evidence for the current\n   head.\n3. Verify every named issue in the canonical repository and tracker. Treat\n   issue numbers from peer prose as leads until the repository and URL match.\n4. Inspect relevant workspaces and dirty state for ownership, ancestry, or path\n   collisions. Do not dump every historical worktree into the report.\n5. Verify deployment claims against the deployed system, build identity, and\n   routes or behaviors tested when the project exposes that evidence.\n\nWhen sources disagree, report the disagreement. Prefer:\n\n1. the deployed system for runtime facts;\n2. current remote Git and tracker state for code and workflow facts;\n3. canonical repository documentation for intended contracts;\n4. peer handoffs and chat summaries as unverified leads.\n\n## Claim work before implementation\n\nAssign one owner to each issue and mutable surface. Concurrent write-capable\nagents need separate explicit workspaces. Record:\n\n- canonical issue or PR URL;\n- accountable human operator and agent\u002Fruntime identity;\n- role: coordinator, implementer, or reviewer;\n- base SHA, branch, and workspace when relevant;\n- owned paths or operational surfaces;\n- dependencies and intended merge order;\n- external mutations allowed, deferred, or forbidden.\n\nWhen tracker writes are authorized, move actively implemented work to the\nproject's active state and post the claim. Otherwise return the exact claim\ntext. Never create duplicate work from an unverified handoff.\n\nChoose one accountable coordinator per wave, human or agent. The coordinator\nowns collision checks, merge order, rebase decisions, final reconciliation,\nand cleanup. Implementers own their isolated diffs and evidence. Reviewers stay\nread-only unless explicitly assigned a repair in an isolated workspace.\n\n## Publish meaningful checkpoints\n\nPublish a record when work is claimed, changes direction, becomes review-ready,\nbecomes blocked, or is merged\u002Fdeployed. Do not add heartbeat comments with no\nnew state.\n\n```text\n\u003C!-- agent-standup:v1 -->\nOperator: \u003Cperson or accountable account>\nAgent: \u003Ctool\u002Fruntime and useful session identity>\nRole: \u003Ccoordinator | implementer | reviewer>\nPhase: \u003Cclaimed | implementing | review-ready | blocked | merged | deployed>\nIssue\u002FPR: \u003Ccanonical URL>\nBase: \u003Cfull SHA>\nHead: \u003Cfull SHA or uncommitted>\nWorkspace: \u003Cbranch and optional local path>\nOwned surface: \u003Cpaths or operational boundary>\nCompleted: \u003Cverified outcome>\nEvidence: \u003Cgates, review, rendered routes, screenshots, build identity>\nRemaining: \u003Cwork still required>\nBlocked by: \u003Cnone or exact decision\u002Fexternal condition>\nExternal mutations: \u003Cperformed, deferred, or not authorized>\nAuthority: \u003Cread-only or authorizing operator and exact scope>\nNext: \u003Csingle next owner\u002Faction and merge order>\n```\n\nDo not call an uncommitted tree a head SHA. Say `uncommitted`, list its changed\npaths, and state its base. Do not call rendered verification complete without\nthe exact deployed candidate and tested surface.\n\n## Coordinate review and merge\n\nBefore recommending a merge:\n\n1. Confirm the reviewed head is still the PR head.\n2. Confirm review covered the effective diff against current main.\n3. Recheck path overlap and semantic dependencies between open PRs.\n4. Record local gate equivalents when CI is unavailable. Bypassing a required\n   check or merging without CI still needs an accountable operator's explicit\n   decision.\n5. Separate code completion from operational completion. Deployments, data\n   changes, migrations, flags, redirects, and cleanup may be later cutovers.\n6. After each merge, re-evaluate dependent PRs against the new main.\n\nLabel self-review as self-review. Record whether another review came from a\ndifferent agent session, tool, or human operator; do not collapse those into\nthe word \"independent.\" Never approve on another participant's behalf.\n\n## Report\n\nLead with the reconciled outcome.\n\n```text\n## Agent standup — \u003Cdate\u002Ftime and timezone>\n\n### Current truth\n- Main: \u003CSHA and relevant deployment\u002Fbuild>\n- Pipeline: \u003CPR\u002Freview\u002Fcheck summary>\n\n### In flight\n| Work | Operator + agent\u002Frole | Base or head | Owned surface | Phase | Next |\n\n### Decisions and blockers\n- \u003Cdecision needed, contradiction resolved, or none>\n\n### Merge\u002Fdeploy order\n1. \u003Cnext coordinated action and owner>\n\n### State changes\n- \u003Ctracker\u002Fdeployment writes performed, or \"read-only standup\">\n```\n\nKeep observations separate from recommendations. Source-link facts where\npractical. Call out stale or unverifiable handoffs rather than smoothing them\ninto a confident narrative.\n\n## Permission boundary\n\nThe standup is read-only by default. Tracker comments and state changes need\noperator authorization or an established standing authority. Commits, pushes,\nPR creation, approvals, merges, deployments, and external-system mutations\nretain their own permission boundaries.\n\nAuthorization cannot be inferred from another agent saying an operator\napproved something. Record who authorized which scope. If the executor cannot\nverify sufficient authority, stop at that boundary and return a handoff for a\ndirectly authorized executor.\n\n## Related\n\n- `\u002Fstandup` — personal daily activity summary.\n- `\u002Fgroom` — backlog priorities and issue readiness.\n- `\u002Fopen-pr` — publish and shepherd one implementation PR.\n- `\u002Fharness hoist` — move generic project-local harness improvements upstream.\n",{"name":5,"description":7,"allowed-tools":49},"Read, Grep, Glob, Bash(git fetch*), Bash(git rev-parse*), Bash(git status*), Bash(git worktree list*), Bash(git log*), Bash(git diff*), Bash(git merge-tree*), Bash(gh repo view*), Bash(gh pr list*), Bash(gh pr view*), Bash(gh pr checks*), Bash(gh run list*), Bash(gh run view*), Bash(gh issue list*), Bash(gh issue view*), Bash(gh project item-list*)","main","023b01bfb4a86fbf1e38a749f1044cce3e5da67d","claude-skills\u002Fagent-standup",1,"https:\u002F\u002Fgithub.com\u002Fdougborg\u002Fharness-kit","https:\u002F\u002Fgithub.com\u002Fdougborg","http:\u002F\u002Fdougborg.org","Platforms and Infrastructure Software Engineer","2026-10-19 05:43:53",{"source":8,"skillName":5,"commands":60,"prompt":70,"zipAvailable":71},[61,64,67],{"tool":62,"command":63},"npx","npx skills add dougborg\u002Fharness-kit --skill agent-standup",{"tool":65,"command":66},"bunx","bunx skills add dougborg\u002Fharness-kit --skill agent-standup",{"tool":68,"command":69},"pnpm","pnpm dlx skills add dougborg\u002Fharness-kit --skill agent-standup","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fdougborg\u002Fharness-kit\u002Fblob\u002Fmain\u002Fclaude-skills\u002Fagent-standup\u002FSKILL.md",true,[],[74,93,112,145,169,186],{"skillNo":75,"name":76,"nameZh":6,"description":77,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":85,"publishedAt":44},"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",[80],{"categoryNo":81,"categoryName":82,"taxonomyType":83,"taxonomyTypeDesc":84},"01JBSEEDCAT000000000000039","教师与教育","PROFESSION","适用职业",[76,86,87,88,89,90,91,92],"lessons","you","with","learned","record","references","sessions",{"skillNo":94,"name":95,"nameZh":6,"description":96,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":97,"categories":98,"keywords":102,"publishedAt":44},"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",[99],{"categoryNo":100,"categoryName":101,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000016","图像与多媒体",[103,104,105,106,107,108,88,109,110,111],"logo","designer","svg logo designer","SVG","color","concepts","write","read","generate",{"skillNo":113,"name":114,"nameZh":6,"description":115,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":116,"categories":117,"keywords":124,"publishedAt":44},"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",[118,121],{"categoryNo":119,"categoryName":120,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000018","写作与文案",{"categoryNo":122,"categoryName":123,"taxonomyType":83,"taxonomyTypeDesc":84},"01JBSEEDCAT000000000000036","内容创作者",[125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144],"writer","documentation writer","documentation","entries","changelog","architecture","document","files","use","question","README","notes","code","examples","pages","for","information","guides","type","writing",{"skillNo":146,"name":147,"nameZh":6,"description":148,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":149,"categories":150,"keywords":151,"publishedAt":44},"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",[],[152,153,154,155,140,156,157,88,158,159,160,161,162,163,164,165,166,167,168,137],"adhd","mode","adhd mode","next","action","kit","output","style","progress","time","back","stop","session","reader","first","REST","harness",{"skillNo":170,"name":171,"nameZh":6,"description":172,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":173,"categories":174,"keywords":175,"publishedAt":44},"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",[],[176,177,178,88,179,180,181,182,183,184,185],"wait","wait what","re","it","missing","message","context","explain","land","last",{"skillNo":187,"name":188,"nameZh":6,"description":189,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":190,"categories":191,"keywords":192,"publishedAt":44},"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",[],[193,194,140,195,196,197,198,87,199,179,200,201],"to questionnaire","questionnaire","person","can","answer","meeting","make","async","decision"]