[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M74PJ2QA2GD2GKNC6G5QFEAF":3,"skill-same:01M74PJ2QA2GD2GKNC6G5QFEAF":50},{"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":28,"commitSha":29,"skillDirectory":30,"forksCount":31,"repositoryHtmlUrl":32,"repositoryHomepageUrl":33,"ownerHtmlUrl":34,"ownerBlogUrl":35,"ownerDescription":6,"updatedAt":36,"install":37,"resources":49},"01M74PJ2QA2GD2GKNC6G5QFEAF","split-plan-tasks",null,"Split a plan into PR-sized groups of small, reviewable tasks appended to the plan file.","eai-org\u002Fagent-toolkit","eai-org","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F301886321?v=4","01M74PFC3HF57T58MQ14P212BQ","Engineering in the age of AI","01M74PFC46MSKTT9SW5Q437FSD",85,"MIT","https:\u002F\u002Fgithub.com\u002Feai-org\u002Fagent-toolkit\u002Fblob\u002Fmain\u002Fskills\u002Fsplit-plan-tasks\u002FSKILL.md",[],[],"2026-11-11 15:11:14",false,"---\nname: split-plan-tasks\ndescription: Split a plan into PR-sized groups of small, reviewable tasks appended to the plan file.\ndisable-model-invocation: true\ntype: flow\nlicense: MIT\nmetadata:\n  version: \"0.6\"\n---\n\n# Split plan into tasks\n\nAppend a task breakdown to a plan so the work lands in small increments — one task, then a review,\na correction pass, a commit — instead of one sweep. Authoring only: executing the tasks is a later,\nseparate session.\n\nInvocation: `\u002Fsplit-plan-tasks \u003Cplan-path>`\n\n## Read the plan\n\nAmbiguous or missing path → ask. A plan with numbered steps and an acceptance-criteria checklist\nsplits best; any plan markdown works — reference tasks to headings when steps aren't numbered, and\ndrop the AC line when there is no such checklist.\n\nA plan already carrying a task section → **stop and ask** which the user wants: replace it, add a\nsecond one beside it, or carry the ticked tasks forward and reformulate only the remaining work.\n\nDone when the plan is read and, where a section already existed, the user has chosen.\n\n## Work out the split\n\n- **Task** — the smallest chunk that leaves the project building and its tests passing, so it can be\n  reviewed and committed on its own. Steps that only make sense together are one task, never two —\n  building alone isn't the test: a step that only re-runs a generator (API clients, specs,\n  snapshots) lands with the change it regenerates, and a step whose test must fail before the fix\n  is written lands with that fix.\n- **Group** — one PR: independently mergeable, delivering something coherent. Order the groups and\n  state each one's dependency on earlier groups.\n- Respect the plan's ordering: a task never precedes what it depends on.\n- **Front-load a hands-on slice.** Among dependency-legal orderings, prefer the one whose first\n  group leaves the app exercisable by hand — a screen to click where the plan touches a UI, else a\n  command to run or an endpoint to hit — accepting a larger first group when that's what makes the\n  slice whole, never scaffolding the plan doesn't call for. When nothing early gets there, say so\n  when presenting.\n- Both ways: every plan step lands in exactly one task and every acceptance criterion in at least\n  one, and no task invents work the plan doesn't call for.\n\nPresent the whole breakdown with the obvious calls already decided, and walk only the genuinely\ndoubtful boundaries as individual questions, each carrying your recommendation, worded via\n`explain-in-simple-language` when available. One approval gate before writing anything.\n\nDone when every step and criterion is placed, the first group is exercisable or you've said why\nnot, and the user has approved the breakdown.\n\n## Append the section\n\nAt the end of the plan file — the only edit made:\n\n```markdown\n## Tasks\n\n### Group 1 — \u003Cwhat this PR delivers>\n\n- *Depends on* nothing\n- *Try it* \u003Cwhat to open or run to exercise the app>\n\n- [ ] **1.1 \u003Cshort imperative>** — steps 2–3\n  - *Done when* \u003Cwhat to run or look at to confirm it landed>\n  - *Satisfies* AC-1, AC-4\n  - *Commit* `\u003Csubject in the repo's commit style>`\n```\n\n*Try it* goes only on the first group that leaves the app exercisable.\n\nDone when every task from the approved breakdown is in the file.\n\n## Close\n\nState the plan's project-relative path and the tally (tasks, groups). Then hand off as a single\ncopy-pasteable launch command — session name and prompt combined, in the launch syntax of the agent\ntool in use (vendor-agnostic — `claude` below is only the example), naming the session\n`execute-tasks-\u003Cslug>`, `\u003Cslug>` being the plan filename without id prefix or extension:\n\n```\nclaude --name execute-tasks-\u003Cslug> \"\u002Fexecute-plan-tasks \u003Cplan-path>\"\n```\n\nThen offer the alternative — clearing the current session instead (vendor-agnostic — `\u002Fclear` is\nonly the example):\n\nOR \u002Fclear and run:\n\n```\n\u002Fexecute-plan-tasks \u003Cplan-path>\n```\n\nDone when the path, the tally, and both commands are printed.\n\n## Boundaries\n\n- The task section is the only edit, replaced only when the user chose so in Read the plan; never\n  touch the plan above it, never touch source files.\n- No implementation — the breakdown is the output.\n","# Split plan into tasks\n\nAppend a task breakdown to a plan so the work lands in small increments — one task, then a review,\na correction pass, a commit — instead of one sweep. Authoring only: executing the tasks is a later,\nseparate session.\n\nInvocation: `\u002Fsplit-plan-tasks \u003Cplan-path>`\n\n## Read the plan\n\nAmbiguous or missing path → ask. A plan with numbered steps and an acceptance-criteria checklist\nsplits best; any plan markdown works — reference tasks to headings when steps aren't numbered, and\ndrop the AC line when there is no such checklist.\n\nA plan already carrying a task section → **stop and ask** which the user wants: replace it, add a\nsecond one beside it, or carry the ticked tasks forward and reformulate only the remaining work.\n\nDone when the plan is read and, where a section already existed, the user has chosen.\n\n## Work out the split\n\n- **Task** — the smallest chunk that leaves the project building and its tests passing, so it can be\n  reviewed and committed on its own. Steps that only make sense together are one task, never two —\n  building alone isn't the test: a step that only re-runs a generator (API clients, specs,\n  snapshots) lands with the change it regenerates, and a step whose test must fail before the fix\n  is written lands with that fix.\n- **Group** — one PR: independently mergeable, delivering something coherent. Order the groups and\n  state each one's dependency on earlier groups.\n- Respect the plan's ordering: a task never precedes what it depends on.\n- **Front-load a hands-on slice.** Among dependency-legal orderings, prefer the one whose first\n  group leaves the app exercisable by hand — a screen to click where the plan touches a UI, else a\n  command to run or an endpoint to hit — accepting a larger first group when that's what makes the\n  slice whole, never scaffolding the plan doesn't call for. When nothing early gets there, say so\n  when presenting.\n- Both ways: every plan step lands in exactly one task and every acceptance criterion in at least\n  one, and no task invents work the plan doesn't call for.\n\nPresent the whole breakdown with the obvious calls already decided, and walk only the genuinely\ndoubtful boundaries as individual questions, each carrying your recommendation, worded via\n`explain-in-simple-language` when available. One approval gate before writing anything.\n\nDone when every step and criterion is placed, the first group is exercisable or you've said why\nnot, and the user has approved the breakdown.\n\n## Append the section\n\nAt the end of the plan file — the only edit made:\n\n```markdown\n## Tasks\n\n### Group 1 — \u003Cwhat this PR delivers>\n\n- *Depends on* nothing\n- *Try it* \u003Cwhat to open or run to exercise the app>\n\n- [ ] **1.1 \u003Cshort imperative>** — steps 2–3\n  - *Done when* \u003Cwhat to run or look at to confirm it landed>\n  - *Satisfies* AC-1, AC-4\n  - *Commit* `\u003Csubject in the repo's commit style>`\n```\n\n*Try it* goes only on the first group that leaves the app exercisable.\n\nDone when every task from the approved breakdown is in the file.\n\n## Close\n\nState the plan's project-relative path and the tally (tasks, groups). Then hand off as a single\ncopy-pasteable launch command — session name and prompt combined, in the launch syntax of the agent\ntool in use (vendor-agnostic — `claude` below is only the example), naming the session\n`execute-tasks-\u003Cslug>`, `\u003Cslug>` being the plan filename without id prefix or extension:\n\n```\nclaude --name execute-tasks-\u003Cslug> \"\u002Fexecute-plan-tasks \u003Cplan-path>\"\n```\n\nThen offer the alternative — clearing the current session instead (vendor-agnostic — `\u002Fclear` is\nonly the example):\n\nOR \u002Fclear and run:\n\n```\n\u002Fexecute-plan-tasks \u003Cplan-path>\n```\n\nDone when the path, the tally, and both commands are printed.\n\n## Boundaries\n\n- The task section is the only edit, replaced only when the user chose so in Read the plan; never\n  touch the plan above it, never touch source files.\n- No implementation — the breakdown is the output.\n",{"name":5,"type":24,"license":15,"metadata":25,"description":7,"disable-model-invocation":27},"flow",{"version":26},"0.6",true,"main","12db59437c9d48b9bd62b18888e9974b17adafaf","skills\u002Fsplit-plan-tasks",19,"https:\u002F\u002Fgithub.com\u002Feai-org\u002Fagent-toolkit","https:\u002F\u002Feai-org.github.io\u002Fagent-toolkit-docs","https:\u002F\u002Fgithub.com\u002Feai-org","https:\u002F\u002Fmedium.com\u002Fengineering-in-the-age-of-ai","2026-11-14 22:36:21",{"source":8,"skillName":5,"commands":38,"prompt":48,"zipAvailable":27},[39,42,45],{"tool":40,"command":41},"npx","npx skills add eai-org\u002Fagent-toolkit --skill split-plan-tasks",{"tool":43,"command":44},"bunx","bunx skills add eai-org\u002Fagent-toolkit --skill split-plan-tasks",{"tool":46,"command":47},"pnpm","pnpm dlx skills add eai-org\u002Fagent-toolkit --skill split-plan-tasks","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Feai-org\u002Fagent-toolkit\u002Fblob\u002Fmain\u002Fskills\u002Fsplit-plan-tasks\u002FSKILL.md",[],[51,57,64,71,78,85],{"skillNo":52,"name":53,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":54,"categories":55,"keywords":56,"publishedAt":19},"01M74PKAVPTTY5PSSAWHCRVY9T","no-frontmatter","https:\u002F\u002Fgithub.com\u002Feai-org\u002Fagent-toolkit\u002Fblob\u002Fmain\u002Ftest\u002Fcompact-skill-creator\u002Ffixtures\u002Fno-frontmatter\u002FSKILL.md",[],[],{"skillNo":58,"name":59,"nameZh":6,"description":60,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":61,"categories":62,"keywords":63,"publishedAt":19},"01M74PK8J99W7V3669Y4JG2JKV","other-name","Fixture skill whose name differs from its folder.","https:\u002F\u002Fgithub.com\u002Feai-org\u002Fagent-toolkit\u002Fblob\u002Fmain\u002Ftest\u002Fcompact-skill-creator\u002Ffixtures\u002Fname-mismatch\u002FSKILL.md",[],[],{"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":70,"publishedAt":19},"01M74PK6T9DHVRCNHTEV2ZA4Y3","md-off","A fixture skill whose folder turns the line length off below a root that sets 99.","https:\u002F\u002Fgithub.com\u002Feai-org\u002Fagent-toolkit\u002Fblob\u002Fmain\u002Ftest\u002Fcompact-skill-creator\u002Ffixtures\u002Fmd-off\u002FSKILL.md",[],[],{"skillNo":72,"name":73,"nameZh":6,"description":74,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":75,"categories":76,"keywords":77,"publishedAt":19},"01M74PK4H4N3D290V4C08M4DGY","long-line","Fixture skill with one prose line of 100 characters, over the fixtures' limit of 99.","https:\u002F\u002Fgithub.com\u002Feai-org\u002Fagent-toolkit\u002Fblob\u002Fmain\u002Ftest\u002Fcompact-skill-creator\u002Ffixtures\u002Flong-line\u002FSKILL.md",[],[],{"skillNo":79,"name":80,"nameZh":6,"description":81,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":82,"categories":83,"keywords":84,"publishedAt":19},"01M74PK2DJ2MCR5KFPMYVQWNJG","flag-without-flow","Fixture skill with the flag and no type.","https:\u002F\u002Fgithub.com\u002Feai-org\u002Fagent-toolkit\u002Fblob\u002Fmain\u002Ftest\u002Fcompact-skill-creator\u002Ffixtures\u002Fflag-without-flow\u002FSKILL.md",[],[],{"skillNo":86,"name":87,"nameZh":6,"description":88,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":89,"categories":90,"keywords":91,"publishedAt":19},"01M74PK03XGEBQ986FDAE46WCM","double-backtick","A fixture skill with double-backtick code spans holding a link template and an overrun.","https:\u002F\u002Fgithub.com\u002Feai-org\u002Fagent-toolkit\u002Fblob\u002Fmain\u002Ftest\u002Fcompact-skill-creator\u002Ffixtures\u002Fdouble-backtick\u002FSKILL.md",[],[]]