[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7CP7TWX07ST84HT5XNEYS1J":3,"skill-same:01M7CP7TWX07ST84HT5XNEYS1J":57},{"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":25,"commitSha":26,"skillDirectory":27,"forksCount":28,"repositoryHtmlUrl":29,"repositoryHomepageUrl":30,"ownerHtmlUrl":31,"ownerBlogUrl":6,"ownerDescription":32,"updatedAt":33,"install":34,"resources":46},"01M7CP7TWX07ST84HT5XNEYS1J","orchestrate",null,"Turn the owner's feature or problem into an approved plan across tickets, keep its state in the workbench, and run it to an announced close.","lutzseverino\u002Frepo-canon","lutzseverino","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F28309837?v=4","01M77NN0YKH0YYGY7W7ER23MX0","Jasper","01M7CP5B58PDMK3RT3CE7F70Q2",1,"MIT","https:\u002F\u002Fgithub.com\u002Flutzseverino\u002Frepo-canon\u002Fblob\u002Fmain\u002F.repo-standards\u002Finputs\u002Fsource\u002F.agents\u002Fskills\u002Forchestrate\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: orchestrate\ndescription: Turn the owner's feature or problem into an approved plan across tickets, keep its state in the workbench, and run it to an announced close.\ndisable-model-invocation: true\n---\n\n# Orchestrate\n\nYou are the orchestrating agent. Turn the owner's feature or problem into a\nplan they approve, with a definite goal; keep the plan's state in the\nworkbench; and run the plan to an announced close.\n\nEach repository's contribution guide, `CONTRIBUTING.md`, decides triage,\nreadiness, question routing and delivery for its work: its Triage, Readiness,\nComments, Findings and Pull requests sections. This skill is how to carry\nthose rules across a plan. The installed `triage` skill is how to triage one\nrequest.\n\n## Workbench\n\nThe workbench is a git repository of plain Markdown holding the state of every\nplan you own. Your global agent instructions name its path. When they name\nnone, ask the owner for it before you propose a plan.\n\nCommit each workbench change straight to its default branch, and push it when\nthe workbench has a remote.\n\n```text\nindex.md     the active plans, each with its owning session reference, and launch caps\ninbox.md     the decision inbox\nwishes.md    one line per wish\n\u003Cplan>\u002F      one directory per plan\n  plan.md    the plan\n  *.md       its grill decisions and draft briefs\n```\n\nThe workbench is a wiki:\n\n- Date every entry.\n- Cite each source, such as an issue, pull request, comment or session, by its\n  link or reference, and leave its content where it is.\n- Leave a superseded entry's text as written: mark it superseded, dated, and\n  point to the entry that replaces it.\n\nA session reference is the opaque identifier your harness gives a session,\nrecorded as given.\n\n### `index.md`\n\nOne line per active plan: its directory, its goal in a few words, and the\nreference of the session that owns it. When a plan closes, move its line under\na `Closed` heading, dated. A `Launch caps` list names a host by its hostname and\nthe most agent launches you may run at once on it.\n\n### `inbox.md`\n\nThe decision inbox the contribution guide names: one entry per decision open\nfor the owner, across plans. Each entry holds the question, its plan, a link\nto its source, the date it opened, and your recommendation.\nGrills and triage runs add entries too. When the owner answers, move the entry\nto its plan's log, with the date, the answer, and where the owner gave it.\n\n### `wishes.md`\n\nOne line per wish: the date, the owner's quoted words or idea, a link\nto its source, and the repository it concerns.\n\nA wish may link a filed finding for future planning. A wish never replaces\nthe finding's tracker record or reporting links.\n\n## The plan\n\n`plan.md` takes this shape:\n\n```markdown\n# \u003CPlan title>\n\n**Goal:** \u003Cthe outcome the plan delivers>\n**Done means:** \u003Cthe observable state that ends it>\n**Approval:** \u003Cdate>, \u003Clink to the owner's approval>. Preapproves: \u003Cwhat the plan names that the approval covers>\n\n## Tickets\n\n| Item | Repository | Issue | State |\n| ---- | ---------- | ----- | ----- |\n\n## Open issues\n\n| Repository | At start | At close |\n| ---------- | -------- | -------- |\n\n## Log\n\n- \u003Ctimestamp> \u003Centry>\n```\n\n- **Tickets.** Every row ends with a closed issue in its Issue column or\n  `dropped (owner)` in its State; the plan closes only then. A row stays for\n  the life of the plan: a split adds rows, a merge points several rows at one\n  issue, and only the owner drops one. State holds the row's latest step, such\n  as `grill \u003Csession reference>`, `grilled \u003Cfile>`, `triaged \u003Cfile>`, `filed`,\n  `ready`, or a pull request link. `grilled \u003Cfile>` and `triaged \u003Cfile>` are\n  the done marks a grill and a triage run set.\n- **Open issues.** Count each repository's open issues when the plan starts and\n  again at close. Report the counts; they set no limit.\n- **Log.** Take every timestamp from the clock, in UTC, with\n  `date -u +%Y-%m-%dT%H:%MZ`, never by hand.\n\nA **plan update** is any report to the owner about a plan, the proposal\nincluded. Each one lists the inferred-intent items the owner has not yet seen,\nwith their citations, and shows the decision inbox. Log the items it showed.\n\n## Lifecycle\n\n1. **Propose.** Present the goal, Done means and ticket table to the owner for\n   approval, as a plan update. On approval, write the approval line, with what\n   the approval preapproves, record the open-issue counts at start, and add\n   the plan to `index.md`.\n2. **Intake.** Route each new request as the contribution guide's Triage\n   section says, with the `triage` skill's method. Put your own ideas into the\n   issue you file under a `Suggestions (not decisions)` heading. Give each\n   triage run what the `triage` skill gives a research run, the plan directory,\n   and this task:\n   - write its result to the plan directory, citing the brief or findings it\n     posted by link;\n   - add each decision it leaves open for the owner to `inbox.md`;\n   - mark its row `triaged \u003Cfile>` in `plan.md`;\n   - commit and push its writes, as [Workbench](#workbench) says;\n   - report back, as [Report-back](#report-back) says.\n3. **Grill.** When the guide's inference test cannot settle an intent\n   question, open a grill with the owner in its own session, and record the\n   session's reference in the row's State. Give it what the `triage` skill\n   gives a grill, the plan directory, and this task:\n   - write the owner's decisions and the draft briefs to the plan directory;\n   - add each decision it leaves open for the owner to `inbox.md`;\n   - mark its row `grilled \u003Cfile>` in `plan.md`;\n   - commit and push its writes, as [Workbench](#workbench) says;\n   - report back, as [Report-back](#report-back) says.\n\n   Then decide the split, file the issues and accept the briefs, as the guide\n   says. A draft you set aside for later work stays in the plan directory,\n   marked parked, and takes no row. To take over an owner's grill or\n   prototype, continue in that session.\n4. **New requirements.** Handle a Finding under the contribution guide's\n   Findings section even when it lies outside the goal: first link an existing\n   open issue reporting the finding; otherwise file it as that section says.\n   Continue the plan without waiting for the finding's resolution.\n   A requirement that arrives mid-plan takes one route:\n   - **Row:** it serves the goal and passes the guide's inference test. Add it\n     as a row, log it, and list it in the next plan update.\n   - **Amendment:** it changes the goal or the scope. Propose the amendment;\n     it applies once the owner approves it, recorded as a new approval line\n     that supersedes the old one.\n   - **Wish:** it lies outside the goal and is not a Finding. Add it to\n     `wishes.md`; it needs no issue merely because it is a wish.\n5. **Close.** Once every row has ended, record the open-issue counts at close,\n   move the plan under `Closed` in `index.md`, and announce, as the last plan\n   update:\n   - what shipped and what was dropped;\n   - the inferred-intent items and the wishes the plan added;\n   - every issue the plan filed that is still open, each with its reason;\n   - one suggested next direction, taken from the wishes and the parked items.\n\n## Running the plan\n\n- One long-lived orchestrating agent owns every plan in the workbench. A fresh\n  session reads `index.md`, `inbox.md` and the active plans, and nothing else,\n  before it acts. It then records its own session reference as each active\n  plan's owner in `index.md`, marking the previous reference superseded.\n- Give heavy reads, such as whole session transcripts, long logs and large\n  diffs, to a subagent that returns what the step needs, or read a bounded\n  tail.\n- Run at most three agent launches you started at once, or the cap `index.md`\n  names for this host's hostname. Count every grill, triage run, research run\n  and worker.\n- Workers deliver as the contribution guide's Pull requests section says.\n  When applicable installed rules assign a worker's merge to you, read its\n  accepted brief and PR handoff, then resume `babysit`, including when resuming\n  a plan. After an owner question is resolved, follow the applicable installed\n  continuation rules to resume the worker or that same open PR.\n\n## Report-back\n\nThe file in the plan directory is the contract. A grill or triage run reports\nby writing its result there, setting its done mark in `plan.md`, and\ncommitting and pushing both. A wake-up message to your session is only a\nnudge, and losing it loses nothing: act on the files.\n\nWhen your harness cannot send messages between sessions, re-read the active\nplan directories at each turn, and the owner tells you when a grill is done.\n\n## With T3 Code\n\nOn T3 Code, these tools carry the steps:\n\n- Open grills and other owner sessions with `t3_thread_launch`, and record\n  each one's reference in the plan as `t3:\u003Cid>`.\n- A launched session reports back with one `t3_thread_send` in `auto` mode\n  to the orchestrating thread.\n- Start triage runs, research runs and workers with `delegate_task`, always\n  with its effort option set. Delegated children wake their parent by\n  themselves.\n- Watch each worker's pull request yourself with `watch_pull_request`; a\n  subagent can't watch.\n- While a plan has open T3 work, keep a 15-minute `schedule_task` heartbeat\n  that runs `babysit`'s wait command on the worker pull requests no watch\n  covers.\n","# Orchestrate\n\nYou are the orchestrating agent. Turn the owner's feature or problem into a\nplan they approve, with a definite goal; keep the plan's state in the\nworkbench; and run the plan to an announced close.\n\nEach repository's contribution guide, `CONTRIBUTING.md`, decides triage,\nreadiness, question routing and delivery for its work: its Triage, Readiness,\nComments, Findings and Pull requests sections. This skill is how to carry\nthose rules across a plan. The installed `triage` skill is how to triage one\nrequest.\n\n## Workbench\n\nThe workbench is a git repository of plain Markdown holding the state of every\nplan you own. Your global agent instructions name its path. When they name\nnone, ask the owner for it before you propose a plan.\n\nCommit each workbench change straight to its default branch, and push it when\nthe workbench has a remote.\n\n```text\nindex.md     the active plans, each with its owning session reference, and launch caps\ninbox.md     the decision inbox\nwishes.md    one line per wish\n\u003Cplan>\u002F      one directory per plan\n  plan.md    the plan\n  *.md       its grill decisions and draft briefs\n```\n\nThe workbench is a wiki:\n\n- Date every entry.\n- Cite each source, such as an issue, pull request, comment or session, by its\n  link or reference, and leave its content where it is.\n- Leave a superseded entry's text as written: mark it superseded, dated, and\n  point to the entry that replaces it.\n\nA session reference is the opaque identifier your harness gives a session,\nrecorded as given.\n\n### `index.md`\n\nOne line per active plan: its directory, its goal in a few words, and the\nreference of the session that owns it. When a plan closes, move its line under\na `Closed` heading, dated. A `Launch caps` list names a host by its hostname and\nthe most agent launches you may run at once on it.\n\n### `inbox.md`\n\nThe decision inbox the contribution guide names: one entry per decision open\nfor the owner, across plans. Each entry holds the question, its plan, a link\nto its source, the date it opened, and your recommendation.\nGrills and triage runs add entries too. When the owner answers, move the entry\nto its plan's log, with the date, the answer, and where the owner gave it.\n\n### `wishes.md`\n\nOne line per wish: the date, the owner's quoted words or idea, a link\nto its source, and the repository it concerns.\n\nA wish may link a filed finding for future planning. A wish never replaces\nthe finding's tracker record or reporting links.\n\n## The plan\n\n`plan.md` takes this shape:\n\n```markdown\n# \u003CPlan title>\n\n**Goal:** \u003Cthe outcome the plan delivers>\n**Done means:** \u003Cthe observable state that ends it>\n**Approval:** \u003Cdate>, \u003Clink to the owner's approval>. Preapproves: \u003Cwhat the plan names that the approval covers>\n\n## Tickets\n\n| Item | Repository | Issue | State |\n| ---- | ---------- | ----- | ----- |\n\n## Open issues\n\n| Repository | At start | At close |\n| ---------- | -------- | -------- |\n\n## Log\n\n- \u003Ctimestamp> \u003Centry>\n```\n\n- **Tickets.** Every row ends with a closed issue in its Issue column or\n  `dropped (owner)` in its State; the plan closes only then. A row stays for\n  the life of the plan: a split adds rows, a merge points several rows at one\n  issue, and only the owner drops one. State holds the row's latest step, such\n  as `grill \u003Csession reference>`, `grilled \u003Cfile>`, `triaged \u003Cfile>`, `filed`,\n  `ready`, or a pull request link. `grilled \u003Cfile>` and `triaged \u003Cfile>` are\n  the done marks a grill and a triage run set.\n- **Open issues.** Count each repository's open issues when the plan starts and\n  again at close. Report the counts; they set no limit.\n- **Log.** Take every timestamp from the clock, in UTC, with\n  `date -u +%Y-%m-%dT%H:%MZ`, never by hand.\n\nA **plan update** is any report to the owner about a plan, the proposal\nincluded. Each one lists the inferred-intent items the owner has not yet seen,\nwith their citations, and shows the decision inbox. Log the items it showed.\n\n## Lifecycle\n\n1. **Propose.** Present the goal, Done means and ticket table to the owner for\n   approval, as a plan update. On approval, write the approval line, with what\n   the approval preapproves, record the open-issue counts at start, and add\n   the plan to `index.md`.\n2. **Intake.** Route each new request as the contribution guide's Triage\n   section says, with the `triage` skill's method. Put your own ideas into the\n   issue you file under a `Suggestions (not decisions)` heading. Give each\n   triage run what the `triage` skill gives a research run, the plan directory,\n   and this task:\n   - write its result to the plan directory, citing the brief or findings it\n     posted by link;\n   - add each decision it leaves open for the owner to `inbox.md`;\n   - mark its row `triaged \u003Cfile>` in `plan.md`;\n   - commit and push its writes, as [Workbench](#workbench) says;\n   - report back, as [Report-back](#report-back) says.\n3. **Grill.** When the guide's inference test cannot settle an intent\n   question, open a grill with the owner in its own session, and record the\n   session's reference in the row's State. Give it what the `triage` skill\n   gives a grill, the plan directory, and this task:\n   - write the owner's decisions and the draft briefs to the plan directory;\n   - add each decision it leaves open for the owner to `inbox.md`;\n   - mark its row `grilled \u003Cfile>` in `plan.md`;\n   - commit and push its writes, as [Workbench](#workbench) says;\n   - report back, as [Report-back](#report-back) says.\n\n   Then decide the split, file the issues and accept the briefs, as the guide\n   says. A draft you set aside for later work stays in the plan directory,\n   marked parked, and takes no row. To take over an owner's grill or\n   prototype, continue in that session.\n4. **New requirements.** Handle a Finding under the contribution guide's\n   Findings section even when it lies outside the goal: first link an existing\n   open issue reporting the finding; otherwise file it as that section says.\n   Continue the plan without waiting for the finding's resolution.\n   A requirement that arrives mid-plan takes one route:\n   - **Row:** it serves the goal and passes the guide's inference test. Add it\n     as a row, log it, and list it in the next plan update.\n   - **Amendment:** it changes the goal or the scope. Propose the amendment;\n     it applies once the owner approves it, recorded as a new approval line\n     that supersedes the old one.\n   - **Wish:** it lies outside the goal and is not a Finding. Add it to\n     `wishes.md`; it needs no issue merely because it is a wish.\n5. **Close.** Once every row has ended, record the open-issue counts at close,\n   move the plan under `Closed` in `index.md`, and announce, as the last plan\n   update:\n   - what shipped and what was dropped;\n   - the inferred-intent items and the wishes the plan added;\n   - every issue the plan filed that is still open, each with its reason;\n   - one suggested next direction, taken from the wishes and the parked items.\n\n## Running the plan\n\n- One long-lived orchestrating agent owns every plan in the workbench. A fresh\n  session reads `index.md`, `inbox.md` and the active plans, and nothing else,\n  before it acts. It then records its own session reference as each active\n  plan's owner in `index.md`, marking the previous reference superseded.\n- Give heavy reads, such as whole session transcripts, long logs and large\n  diffs, to a subagent that returns what the step needs, or read a bounded\n  tail.\n- Run at most three agent launches you started at once, or the cap `index.md`\n  names for this host's hostname. Count every grill, triage run, research run\n  and worker.\n- Workers deliver as the contribution guide's Pull requests section says.\n  When applicable installed rules assign a worker's merge to you, read its\n  accepted brief and PR handoff, then resume `babysit`, including when resuming\n  a plan. After an owner question is resolved, follow the applicable installed\n  continuation rules to resume the worker or that same open PR.\n\n## Report-back\n\nThe file in the plan directory is the contract. A grill or triage run reports\nby writing its result there, setting its done mark in `plan.md`, and\ncommitting and pushing both. A wake-up message to your session is only a\nnudge, and losing it loses nothing: act on the files.\n\nWhen your harness cannot send messages between sessions, re-read the active\nplan directories at each turn, and the owner tells you when a grill is done.\n\n## With T3 Code\n\nOn T3 Code, these tools carry the steps:\n\n- Open grills and other owner sessions with `t3_thread_launch`, and record\n  each one's reference in the plan as `t3:\u003Cid>`.\n- A launched session reports back with one `t3_thread_send` in `auto` mode\n  to the orchestrating thread.\n- Start triage runs, research runs and workers with `delegate_task`, always\n  with its effort option set. Delegated children wake their parent by\n  themselves.\n- Watch each worker's pull request yourself with `watch_pull_request`; a\n  subagent can't watch.\n- While a plan has open T3 work, keep a 15-minute `schedule_task` heartbeat\n  that runs `babysit`'s wait command on the worker pull requests no watch\n  covers.\n",{"name":5,"description":7,"disable-model-invocation":24},true,"main","f5a3e334f7bc0fce1854f224a8e78c53f8ea75b9",".repo-standards\u002Finputs\u002Fsource\u002F.agents\u002Fskills\u002Forchestrate",0,"https:\u002F\u002Fgithub.com\u002Flutzseverino\u002Frepo-canon","https:\u002F\u002Frepo-standards.com\u002Fsources\u002Flutzseverino\u002Frepo-canon\u002F","https:\u002F\u002Fgithub.com\u002Flutzseverino","If Codex could write profile bios, you wouldn't be reading this. Or would you?","2026-11-15 06:12:31",{"source":8,"skillName":5,"commands":35,"prompt":45,"zipAvailable":24},[36,39,42],{"tool":37,"command":38},"npx","npx skills add lutzseverino\u002Frepo-canon --skill orchestrate",{"tool":40,"command":41},"bunx","bunx skills add lutzseverino\u002Frepo-canon --skill orchestrate",{"tool":43,"command":44},"pnpm","pnpm dlx skills add lutzseverino\u002Frepo-canon --skill orchestrate","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Flutzseverino\u002Frepo-canon\u002Fblob\u002Fmain\u002F.repo-standards\u002Finputs\u002Fsource\u002F.agents\u002Fskills\u002Forchestrate\u002FSKILL.md",[47,53],{"path":48,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":52},"agents\u002Fopenai.yaml","REFERENCE","参考文档",143,"https:\u002F\u002Fgithub.com\u002Flutzseverino\u002Frepo-canon\u002Fblob\u002Fmain\u002F.repo-standards\u002Finputs\u002Fsource\u002F.agents\u002Fskills\u002Forchestrate\u002Fagents\u002Fopenai.yaml",{"path":54,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":55,"sourceHtmlUrl":56},"CREDITS.md",574,"https:\u002F\u002Fgithub.com\u002Flutzseverino\u002Frepo-canon\u002Fblob\u002Fmain\u002F.repo-standards\u002Finputs\u002Fsource\u002F.agents\u002Fskills\u002Forchestrate\u002FCREDITS.md",[58,65,72,79,86,93],{"skillNo":59,"name":60,"nameZh":6,"description":61,"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},"01M7CP829W88YVACQ3ZFS4KYAK","repository-standards","Look up installed rules before any work in a disposable planning exercise, including tracker work and file changes.","https:\u002F\u002Fgithub.com\u002Flutzseverino\u002Frepo-canon\u002Fblob\u002Fmain\u002Fscripts\u002Fsupport\u002Frepository-standards\u002FSKILL.md",[],[],{"skillNo":66,"name":67,"nameZh":6,"description":68,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":19},"01M7CP7Y70KNBJBVDNGSQ6RGG4","triage","Move issues and external PRs through a state machine of triage roles, sort each request by where its uncertainty lies, and bring its agent brief to acceptance through research and independent review, or through a grilling session with the owner.","https:\u002F\u002Fgithub.com\u002Flutzseverino\u002Frepo-canon\u002Fblob\u002Fmain\u002F.repo-standards\u002Finputs\u002Fsource\u002F.agents\u002Fskills\u002Ftriage\u002FSKILL.md",[],[],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":19},"01M7CP7QE6M9W0E24BV0VZV1RS","babysit","Babysit an opened pull request through checks and review to merge or its prescribed handoff. Use right after opening a pull request, or when asked to babysit or land one.","https:\u002F\u002Fgithub.com\u002Flutzseverino\u002Frepo-canon\u002Fblob\u002Fmain\u002F.repo-standards\u002Finputs\u002Fsource\u002F.agents\u002Fskills\u002Fbabysit\u002FSKILL.md",[],[],{"skillNo":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"publishedAt":19},"01M7CP7N4VMT6C4THVN2RBRB5Q","writing-for-agents","Writing documents for agents. Use when creating or editing skills, or modifying AGENTS.md or CLAUDE.md.","https:\u002F\u002Fgithub.com\u002Flutzseverino\u002Frepo-canon\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fwriting-for-agents\u002FSKILL.md",[],[],{"skillNo":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"publishedAt":19},"01M7CP7GN4YYS18YS91N0AN0PK","wayfinder","Plan a huge chunk of work (more than one agent session can hold) as a shared map of decision tickets on your issue tracker, and resolve them one at a time until the way to the destination is clear.","https:\u002F\u002Fgithub.com\u002Flutzseverino\u002Frepo-canon\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fwayfinder\u002FSKILL.md",[],[],{"skillNo":94,"name":67,"nameZh":6,"description":95,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":96,"categories":97,"keywords":98,"publishedAt":19},"01M7CP7CP9YAQFRV4AC1Q3HFTP","Move issues and external PRs through a state machine of triage roles, sort each request by where its uncertainty lies, and bring its agent brief to acceptance through research and independent review, or through a grilling session with the owner. Also runs unattended as the factory's triage run, which writes nothing and ends with a run result.","https:\u002F\u002Fgithub.com\u002Flutzseverino\u002Frepo-canon\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Ftriage\u002FSKILL.md",[],[]]