[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6DZRV3XD70JM27G63JRHG44":3,"skill-same:01M6DZRV3XD70JM27G63JRHG44":110},{"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":49,"commitSha":50,"skillDirectory":51,"forksCount":52,"repositoryHtmlUrl":53,"repositoryHomepageUrl":54,"ownerHtmlUrl":55,"ownerBlogUrl":56,"ownerDescription":6,"updatedAt":57,"install":58,"resources":71},"01M6DZRV3XD70JM27G63JRHG44","orbit-orchestrate",null,"Prepares and supervises an Orbit backlog. Deduplicates and pilots tasks (task-pilot), assigns crews, promotes ready work, dispatches with `orbit run ship` and `orbit run auto` drains, diagnoses failed or stuck `jrun-*` runs, and routes CI\u002FQA sweep findings into repair tasks. Use for delivery operations across tasks, jobs, activities and workflow runs. Use orbit for one assigned task and orbit-setup for machine configuration or scheduler installation.","constellation-works\u002Forbit","constellation-works","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F322971908?v=4","01M6DZR7HBGAB3RYGA69FES9BR","Constellation-Works","01M6DZR7NKYTEFH424SK4BHCX3",21,"MIT","https:\u002F\u002Fgithub.com\u002Fconstellation-works\u002Forbit\u002Fblob\u002Fagent-main\u002Fplugin\u002Fskills\u002Forbit-orchestrate\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],"orbit orchestrate","orbit","orchestrate","machine","qa sweep","for","backlog","repair","ship","scheduler","delivery","run","work","across","tasks","workflow","auto","qa","ci","ready","2026-11-10 10:12:30",false,"---\nname: orbit-orchestrate\ndescription: Prepares and supervises an Orbit backlog. Deduplicates and pilots tasks (task-pilot), assigns crews, promotes ready work, dispatches with `orbit run ship` and `orbit run auto` drains, diagnoses failed or stuck `jrun-*` runs, and routes CI\u002FQA sweep findings into repair tasks. Use for delivery operations across tasks, jobs, activities and workflow runs. Use orbit for one assigned task and orbit-setup for machine configuration or scheduler installation.\n---\n\n# Orbit Orchestrate\n\nKeep authorized work moving using durable task, run and delivery evidence.\nA managed implementation worker stays within its assigned task; this skill\nis for the operator supervising work.\n\n## Operating loop\n\nInspect → deduplicate → author → prepare → promote → dispatch → verify → repair.\n\n- Discover the owning workspace and use its returned selector; see\n  [tool surface](..\u002Forbit\u002Freferences\u002Ftool-surface.md). Never substitute a local\n  store for an unavailable owner.\n- Follow the user's scope, crew choices and concurrency\u002Fbuild budget. After\n  authorized preparation succeeds, promote promptly. An active drain can claim\n  backlog work immediately, so finish scope changes before exposing it.\n- Creation, promotion and completion are distinct. `--complete` authorizes\n  delivery for work admitted in that invocation's window; it does not itself\n  authorize promotion, future windows or releases; `run auto --approve-proposed`\n  is the separate opt-in for a drain to approve qualifying proposed tasks, and\n  `run auto --auto-fix` the separate opt-in to repair its own PRs' failed\n  required checks and conflicts, which never merges.\n  Existing continuous-delivery authorization may cover them; apply the user's\n  actual instructions.\n- Verify persisted preparation, actual diffs, required checks and merge state.\n  Post-merge review\u002FQA findings become repairs under continuous delivery; do not\n  invent an extra pre-merge approval gate. Repository protections still apply.\n- Start every supervision pass (each wake of a long-running session) with\n  `orbit.attention` (`orbit attention`) on the owning host for the workspaces\n  you supervise. Clear each `pilot_warning` and `pilot_hold`: repair the task\n  (criteria, context, dependencies; `required_tools` is set only at filing, so\n  re-file to change it) and re-pilot, or record on the task why it stays.\n  Treat an empty result as clear only when `degraded` is empty and the read\n  reached the owner: a federated read that misses the owner can return\n  nothing. A caller holding neither the agent nor the operator capability\n  is refused with `capability_denied` and does not receive an empty\n  attention envelope.\n- Diagnose blocked work before retrying. Preserve user interventions and active\n  work. A stopped admission loop does not mean its children were cancelled.\n- Record `orchestrator` separately from execution `crew`. Neither attribution\n  field grants authority. Leave a durable handoff when a lane cannot proceed.\n\n## Mid-run steering\n\nUse a steering comment to correct the agent's course while the admitted task\nscope and acceptance criteria remain right. Do not cancel a run just to add\nguidance: steering is delivered at the next prompt that includes task context,\nnot injected into a step that is already running.\n\nPost an operator comment on the task while the run is in flight:\n\n```bash\norbit task update \u003Ctask-id> --comment \"Please prioritize the existing acceptance criteria for the retry path.\"\n```\n\nThis also works for a claimed task. Write to the owner's task record; the owner\nadmits an operator's comment-only update while the execution claim is active,\nand the claimed leaf reads and records delivery through that owner. Other\nunscoped task edits remain refused during the claim. Automation and agent\ncomments are not steering.\n\nThe run carries outstanding operator comments, oldest first, in its next safe\ntask-context prompt. Each comment is delivered at most once per run. Verify\ndelivery with `orbit task show \u003Ctask-id>`: find the `steering_delivered` event\nfor the run and step, then confirm the comment is listed under `delivered`. A\ncomment under `truncated` was not carried in full. The host maintains the\ndelivery watermark as internal pipeline state; the event is the operator-facing\nreceipt. Agent narrative in the execution summary may help diagnose a run, but\nis not required to confirm delivery.\n\nWhen steering is enabled, `task_pr_pipeline` and `task_local_pipeline`, plus\n`task_claimed_pr_pipeline` and `task_claimed_local_pipeline`, check for\nundelivered comments before `commit`. An outstanding comment sends the\nuncommitted candidate through a bounded steering round, up to `max_rounds`.\nIf the limit is spent, or the last round delivers none of the remaining\ncomments, the task is held as `steering_unconsumed` and the run stops without\ncommitting. Inspect the task comments, then requeue, reply with clarification,\nor accept the candidate through the normal review record.\n\nConfigure steering for the workspace:\n\n```toml\n[workflow.mid_run_steering]\nenabled = true\nmax_rounds = 2\n```\n\n`enabled` defaults to `false`; `max_rounds` defaults to `2` and accepts\n`1` through `10`. Settings refresh at prompt and gate checkpoints. Enabling\nwhile runs are already in flight leaves missing watermarks absent, so operator\ncomments since dispatch are outstanding at the next prompt or gate unless an\ninherited candidate or claim watermark provides the baseline; prefer enabling\nbetween drains. Disabling takes effect at the next checkpoint and leaves\nexisting watermark and round state inert.\n\nSteering cannot change the task's actual specification. A changed title,\ndescription, or acceptance criteria changes the task meaning digest. If the\nold-scope candidate must not continue, cancel the live child run if needed and\nwait for it and its claim to settle. Then update the task and requeue for fresh\nimplementation with the old candidate discarded. See\n[failed runs](references\u002Forchestration.md#failed-runs) for requeue mechanics.\n\n## Choose the reference\n\n| Need | Read |\n|---|---|\n| Correct course while a task is running | [Mid-run steering](#mid-run-steering) |\n| Prepare\u002Fpromote work and supervise progress | [Operating loop](references\u002Floop.md) |\n| Compare implementer models when selecting a crew | [Model ranking](references\u002Fmodel-ranking.md) |\n| Scope completion authority and dispatch windows | [Authorization](references\u002Fauthorization.md) |\n| Ship\u002Fauto command examples, locks and capacity | [Dispatch mechanics](references\u002Forchestration.md) |\n| Understand the drain card's backlog forecast | [Backlog forecast](references\u002Forchestration.md#backlog-forecast) |\n| See admission's CPU\u002Fmemory forecast, its accuracy and decisions | [Resource forecast view](references\u002Forchestration.md#resource-forecast-view) |\n| Jobs, activities, run commands and CI sweep | [Workflows](references\u002Fworkflows.md) |\n| Author or diagnose version 3 auto-task jobs, waits and accepted results | [Auto-task workflows](references\u002Fworkflows.md#version-3-auto-task-workflows) |\n| Diagnose task-pilot start prerequisites and unreachable evidence | [Task-pilot readiness holds](references\u002Fworkflows.md#task-pilot-readiness-holds) |\n| Diagnose a specific failed, held, or stuck run | [Run debugging](references\u002Frun-debugging.md) |\n| Choose checkpoint resume or a whole-run replay from step zero | [Whole-run replay](references\u002Frun-debugging.md#whole-run-replay-from-step-zero) |\n| Tell a corrected agent reply from a clean one | [Response-correction rounds](references\u002Frun-debugging.md#bounded-response-correction-rounds) |\n| Tell what a held delivery-refusal code means | [Delivery-refusal holds](references\u002Frun-debugging.md#delivery-refusal-holds) |\n| Tell an inherited red base from candidate damage in hosted CI | [Hosted-check diagnosis](references\u002Frun-debugging.md#hosted-check-failures-and-inherited-red-bases) |\n| Diagnose a failed local required-validation command and its replay evidence | [Local validation replay](references\u002Frun-debugging.md#local-required-validation-replay) |\n| Find or answer a human decision request on held work | [Decision requests](references\u002Frun-debugging.md#decision-requests-on-held-work) |\n| Match an observed failure to a remedy | [Common failures](references\u002Fcommon-failures.md) |\n| Route CI findings, recover delivery and verify deployment | [Recovery](references\u002Frecovery.md) |\n| Salvage an incomplete or stopped candidate instead of starting over | [Salvage an incomplete candidate](references\u002Frecovery.md#salvage-an-incomplete-candidate) |\n| Work through typical orchestration decisions | [Walkthroughs](references\u002Fwalkthroughs.md) |\n\nUse [task authoring](..\u002Forbit\u002Freferences\u002Ftask-authoring.md) when filing work.\nUse [orbit-setup](..\u002Forbit-setup\u002FSKILL.md) when the cause is host configuration,\ninstallation, scheduler setup or maintenance; do not load those procedures\nfor ordinary dispatch.\n","# Orbit Orchestrate\n\nKeep authorized work moving using durable task, run and delivery evidence.\nA managed implementation worker stays within its assigned task; this skill\nis for the operator supervising work.\n\n## Operating loop\n\nInspect → deduplicate → author → prepare → promote → dispatch → verify → repair.\n\n- Discover the owning workspace and use its returned selector; see\n  [tool surface](..\u002Forbit\u002Freferences\u002Ftool-surface.md). Never substitute a local\n  store for an unavailable owner.\n- Follow the user's scope, crew choices and concurrency\u002Fbuild budget. After\n  authorized preparation succeeds, promote promptly. An active drain can claim\n  backlog work immediately, so finish scope changes before exposing it.\n- Creation, promotion and completion are distinct. `--complete` authorizes\n  delivery for work admitted in that invocation's window; it does not itself\n  authorize promotion, future windows or releases; `run auto --approve-proposed`\n  is the separate opt-in for a drain to approve qualifying proposed tasks, and\n  `run auto --auto-fix` the separate opt-in to repair its own PRs' failed\n  required checks and conflicts, which never merges.\n  Existing continuous-delivery authorization may cover them; apply the user's\n  actual instructions.\n- Verify persisted preparation, actual diffs, required checks and merge state.\n  Post-merge review\u002FQA findings become repairs under continuous delivery; do not\n  invent an extra pre-merge approval gate. Repository protections still apply.\n- Start every supervision pass (each wake of a long-running session) with\n  `orbit.attention` (`orbit attention`) on the owning host for the workspaces\n  you supervise. Clear each `pilot_warning` and `pilot_hold`: repair the task\n  (criteria, context, dependencies; `required_tools` is set only at filing, so\n  re-file to change it) and re-pilot, or record on the task why it stays.\n  Treat an empty result as clear only when `degraded` is empty and the read\n  reached the owner: a federated read that misses the owner can return\n  nothing. A caller holding neither the agent nor the operator capability\n  is refused with `capability_denied` and does not receive an empty\n  attention envelope.\n- Diagnose blocked work before retrying. Preserve user interventions and active\n  work. A stopped admission loop does not mean its children were cancelled.\n- Record `orchestrator` separately from execution `crew`. Neither attribution\n  field grants authority. Leave a durable handoff when a lane cannot proceed.\n\n## Mid-run steering\n\nUse a steering comment to correct the agent's course while the admitted task\nscope and acceptance criteria remain right. Do not cancel a run just to add\nguidance: steering is delivered at the next prompt that includes task context,\nnot injected into a step that is already running.\n\nPost an operator comment on the task while the run is in flight:\n\n```bash\norbit task update \u003Ctask-id> --comment \"Please prioritize the existing acceptance criteria for the retry path.\"\n```\n\nThis also works for a claimed task. Write to the owner's task record; the owner\nadmits an operator's comment-only update while the execution claim is active,\nand the claimed leaf reads and records delivery through that owner. Other\nunscoped task edits remain refused during the claim. Automation and agent\ncomments are not steering.\n\nThe run carries outstanding operator comments, oldest first, in its next safe\ntask-context prompt. Each comment is delivered at most once per run. Verify\ndelivery with `orbit task show \u003Ctask-id>`: find the `steering_delivered` event\nfor the run and step, then confirm the comment is listed under `delivered`. A\ncomment under `truncated` was not carried in full. The host maintains the\ndelivery watermark as internal pipeline state; the event is the operator-facing\nreceipt. Agent narrative in the execution summary may help diagnose a run, but\nis not required to confirm delivery.\n\nWhen steering is enabled, `task_pr_pipeline` and `task_local_pipeline`, plus\n`task_claimed_pr_pipeline` and `task_claimed_local_pipeline`, check for\nundelivered comments before `commit`. An outstanding comment sends the\nuncommitted candidate through a bounded steering round, up to `max_rounds`.\nIf the limit is spent, or the last round delivers none of the remaining\ncomments, the task is held as `steering_unconsumed` and the run stops without\ncommitting. Inspect the task comments, then requeue, reply with clarification,\nor accept the candidate through the normal review record.\n\nConfigure steering for the workspace:\n\n```toml\n[workflow.mid_run_steering]\nenabled = true\nmax_rounds = 2\n```\n\n`enabled` defaults to `false`; `max_rounds` defaults to `2` and accepts\n`1` through `10`. Settings refresh at prompt and gate checkpoints. Enabling\nwhile runs are already in flight leaves missing watermarks absent, so operator\ncomments since dispatch are outstanding at the next prompt or gate unless an\ninherited candidate or claim watermark provides the baseline; prefer enabling\nbetween drains. Disabling takes effect at the next checkpoint and leaves\nexisting watermark and round state inert.\n\nSteering cannot change the task's actual specification. A changed title,\ndescription, or acceptance criteria changes the task meaning digest. If the\nold-scope candidate must not continue, cancel the live child run if needed and\nwait for it and its claim to settle. Then update the task and requeue for fresh\nimplementation with the old candidate discarded. See\n[failed runs](references\u002Forchestration.md#failed-runs) for requeue mechanics.\n\n## Choose the reference\n\n| Need | Read |\n|---|---|\n| Correct course while a task is running | [Mid-run steering](#mid-run-steering) |\n| Prepare\u002Fpromote work and supervise progress | [Operating loop](references\u002Floop.md) |\n| Compare implementer models when selecting a crew | [Model ranking](references\u002Fmodel-ranking.md) |\n| Scope completion authority and dispatch windows | [Authorization](references\u002Fauthorization.md) |\n| Ship\u002Fauto command examples, locks and capacity | [Dispatch mechanics](references\u002Forchestration.md) |\n| Understand the drain card's backlog forecast | [Backlog forecast](references\u002Forchestration.md#backlog-forecast) |\n| See admission's CPU\u002Fmemory forecast, its accuracy and decisions | [Resource forecast view](references\u002Forchestration.md#resource-forecast-view) |\n| Jobs, activities, run commands and CI sweep | [Workflows](references\u002Fworkflows.md) |\n| Author or diagnose version 3 auto-task jobs, waits and accepted results | [Auto-task workflows](references\u002Fworkflows.md#version-3-auto-task-workflows) |\n| Diagnose task-pilot start prerequisites and unreachable evidence | [Task-pilot readiness holds](references\u002Fworkflows.md#task-pilot-readiness-holds) |\n| Diagnose a specific failed, held, or stuck run | [Run debugging](references\u002Frun-debugging.md) |\n| Choose checkpoint resume or a whole-run replay from step zero | [Whole-run replay](references\u002Frun-debugging.md#whole-run-replay-from-step-zero) |\n| Tell a corrected agent reply from a clean one | [Response-correction rounds](references\u002Frun-debugging.md#bounded-response-correction-rounds) |\n| Tell what a held delivery-refusal code means | [Delivery-refusal holds](references\u002Frun-debugging.md#delivery-refusal-holds) |\n| Tell an inherited red base from candidate damage in hosted CI | [Hosted-check diagnosis](references\u002Frun-debugging.md#hosted-check-failures-and-inherited-red-bases) |\n| Diagnose a failed local required-validation command and its replay evidence | [Local validation replay](references\u002Frun-debugging.md#local-required-validation-replay) |\n| Find or answer a human decision request on held work | [Decision requests](references\u002Frun-debugging.md#decision-requests-on-held-work) |\n| Match an observed failure to a remedy | [Common failures](references\u002Fcommon-failures.md) |\n| Route CI findings, recover delivery and verify deployment | [Recovery](references\u002Frecovery.md) |\n| Salvage an incomplete or stopped candidate instead of starting over | [Salvage an incomplete candidate](references\u002Frecovery.md#salvage-an-incomplete-candidate) |\n| Work through typical orchestration decisions | [Walkthroughs](references\u002Fwalkthroughs.md) |\n\nUse [task authoring](..\u002Forbit\u002Freferences\u002Ftask-authoring.md) when filing work.\nUse [orbit-setup](..\u002Forbit-setup\u002FSKILL.md) when the cause is host configuration,\ninstallation, scheduler setup or maintenance; do not load those procedures\nfor ordinary dispatch.\n",{"name":5,"description":7},"agent-main","ed219d0bda9100437b5f448fe559dfe0eea3b9a4","plugin\u002Fskills\u002Forbit-orchestrate",1,"https:\u002F\u002Fgithub.com\u002Fconstellation-works\u002Forbit","https:\u002F\u002Forbit-cli.com","https:\u002F\u002Fgithub.com\u002Fconstellation-works","https:\u002F\u002Fconstellation-works.com\u002F","2026-11-13 16:45:32",{"source":8,"skillName":5,"commands":59,"prompt":69,"zipAvailable":70},[60,63,66],{"tool":61,"command":62},"npx","npx skills add constellation-works\u002Forbit --skill orbit-orchestrate",{"tool":64,"command":65},"bunx","bunx skills add constellation-works\u002Forbit --skill orbit-orchestrate",{"tool":67,"command":68},"pnpm","pnpm dlx skills add constellation-works\u002Forbit --skill orbit-orchestrate","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fconstellation-works\u002Forbit\u002Fblob\u002Fagent-main\u002Fplugin\u002Fskills\u002Forbit-orchestrate\u002FSKILL.md",true,[72,78,82,86,90,94,98,102,106],{"path":73,"resourceType":74,"resourceTypeDesc":75,"sizeBytes":76,"sourceHtmlUrl":77},"references\u002Fauthorization.md","REFERENCE","参考文档",9848,"https:\u002F\u002Fgithub.com\u002Fconstellation-works\u002Forbit\u002Fblob\u002Fagent-main\u002Fplugin\u002Fskills\u002Forbit-orchestrate\u002Freferences\u002Fauthorization.md",{"path":79,"resourceType":74,"resourceTypeDesc":75,"sizeBytes":80,"sourceHtmlUrl":81},"references\u002Fcommon-failures.md",6804,"https:\u002F\u002Fgithub.com\u002Fconstellation-works\u002Forbit\u002Fblob\u002Fagent-main\u002Fplugin\u002Fskills\u002Forbit-orchestrate\u002Freferences\u002Fcommon-failures.md",{"path":83,"resourceType":74,"resourceTypeDesc":75,"sizeBytes":84,"sourceHtmlUrl":85},"references\u002Floop.md",10759,"https:\u002F\u002Fgithub.com\u002Fconstellation-works\u002Forbit\u002Fblob\u002Fagent-main\u002Fplugin\u002Fskills\u002Forbit-orchestrate\u002Freferences\u002Floop.md",{"path":87,"resourceType":74,"resourceTypeDesc":75,"sizeBytes":88,"sourceHtmlUrl":89},"references\u002Fmodel-ranking.md",12778,"https:\u002F\u002Fgithub.com\u002Fconstellation-works\u002Forbit\u002Fblob\u002Fagent-main\u002Fplugin\u002Fskills\u002Forbit-orchestrate\u002Freferences\u002Fmodel-ranking.md",{"path":91,"resourceType":74,"resourceTypeDesc":75,"sizeBytes":92,"sourceHtmlUrl":93},"references\u002Forchestration.md",67333,"https:\u002F\u002Fgithub.com\u002Fconstellation-works\u002Forbit\u002Fblob\u002Fagent-main\u002Fplugin\u002Fskills\u002Forbit-orchestrate\u002Freferences\u002Forchestration.md",{"path":95,"resourceType":74,"resourceTypeDesc":75,"sizeBytes":96,"sourceHtmlUrl":97},"references\u002Frecovery.md",28647,"https:\u002F\u002Fgithub.com\u002Fconstellation-works\u002Forbit\u002Fblob\u002Fagent-main\u002Fplugin\u002Fskills\u002Forbit-orchestrate\u002Freferences\u002Frecovery.md",{"path":99,"resourceType":74,"resourceTypeDesc":75,"sizeBytes":100,"sourceHtmlUrl":101},"references\u002Frun-debugging.md",53898,"https:\u002F\u002Fgithub.com\u002Fconstellation-works\u002Forbit\u002Fblob\u002Fagent-main\u002Fplugin\u002Fskills\u002Forbit-orchestrate\u002Freferences\u002Frun-debugging.md",{"path":103,"resourceType":74,"resourceTypeDesc":75,"sizeBytes":104,"sourceHtmlUrl":105},"references\u002Fwalkthroughs.md",6121,"https:\u002F\u002Fgithub.com\u002Fconstellation-works\u002Forbit\u002Fblob\u002Fagent-main\u002Fplugin\u002Fskills\u002Forbit-orchestrate\u002Freferences\u002Fwalkthroughs.md",{"path":107,"resourceType":74,"resourceTypeDesc":75,"sizeBytes":108,"sourceHtmlUrl":109},"references\u002Fworkflows.md",53224,"https:\u002F\u002Fgithub.com\u002Fconstellation-works\u002Forbit\u002Fblob\u002Fagent-main\u002Fplugin\u002Fskills\u002Forbit-orchestrate\u002Freferences\u002Fworkflows.md",[111,133,157,163,169,175],{"skillNo":112,"name":25,"nameZh":6,"description":113,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":114,"categories":115,"keywords":117,"publishedAt":44},"01M6DZS3FEK5PCAK0EEXRKDMX6","Creates, executes and reviews Orbit tasks; searches task history and friction records; records evidence. Use for an assigned task or task ID, task authoring, `orbit.task`\u002F`orbit.search` MCP tools, `orbit tool run`, task records under `.orbit\u002F`, or friction. Use orbit-orchestrate for `orbit run` dispatch, drains and failed `jrun-*` runs, and orbit-setup for installing or configuring a machine.","https:\u002F\u002Fgithub.com\u002Fconstellation-works\u002Forbit\u002Fblob\u002Fagent-main\u002Fplugin\u002Fskills\u002Forbit\u002FSKILL.md",[116],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[25,26,118,119,38,35,120,121,122,123,124,125,126,127,128,129,130,131,132,27,29],"search","MCP","authoring","Tools","evidence","history","tool","use","task","id","dispatch","friction","mcp tools","reviews","records",{"skillNo":134,"name":135,"nameZh":6,"description":136,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":137,"categories":138,"keywords":140,"publishedAt":44},"01M6DZRYMNW8BCQ2GGKESTJKX0","orbit-setup","Sets up and maintains Orbit on a user's machine and repositories according to their needs. Use for installation, `orbit workspace init`, onboarding, provider\u002Fcrew and MCP configuration, experimental crew memory, the Linux sandbox, the headless browser kit for UI checks, the scheduler (sweep clock, routines, auto-tasks), worktree GC, `orbit doctor`, upgrades, task publication and backup, remote access, the dashboard, or host and service log troubleshooting. Supports existing installations as well as first-time setup; task execution belongs to orbit and backlog dispatch to orbit-orchestrate.","https:\u002F\u002Fgithub.com\u002Fconstellation-works\u002Forbit\u002Fblob\u002Fagent-main\u002Fplugin\u002Fskills\u002Forbit-setup\u002FSKILL.md",[139],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[141,25,142,26,143,144,145,119,146,40,147,38,148,149,150,151,152,153,154,155,156],"orbit setup","setup","time","up","init","host","access","log","browser","dashboard","gc","worktree","headless","kit","remote","memory",{"skillNo":158,"name":159,"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":160,"categories":161,"keywords":162,"publishedAt":44},"01M6DZRR6KKPQXN3BTPM9HKRB5","graph","https:\u002F\u002Fgithub.com\u002Fconstellation-works\u002Forbit\u002Fblob\u002Fagent-main\u002Fcrates\u002Forbit-tools\u002Ftests\u002Ffixtures\u002Fplugins\u002Fgraph-example\u002F.orbit-plugin\u002Fskills\u002Fgraph\u002FSKILL.md",[],[159],{"skillNo":164,"name":25,"nameZh":6,"description":113,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":165,"categories":166,"keywords":168,"publishedAt":44},"01M6DZRMN4RDBH6YXF6GW5TPVG","https:\u002F\u002Fgithub.com\u002Fconstellation-works\u002Forbit\u002Fblob\u002Fagent-main\u002Fcrates\u002Forbit-core\u002Fassets\u002Fskills\u002Forbit\u002FSKILL.md",[167],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[25,26,118,119,38,35,120,121,122,123,124,125,126,127,128,129,130,131,132,27,29],{"skillNo":170,"name":135,"nameZh":6,"description":136,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":171,"categories":172,"keywords":174,"publishedAt":44},"01M6DZRFM0H5GX49W81JE494CW","https:\u002F\u002Fgithub.com\u002Fconstellation-works\u002Forbit\u002Fblob\u002Fagent-main\u002Fcrates\u002Forbit-core\u002Fassets\u002Fskills\u002Forbit-setup\u002FSKILL.md",[173],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[141,25,142,26,143,144,145,119,146,40,147,38,148,149,150,151,152,153,154,155,156],{"skillNo":176,"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":177,"categories":178,"keywords":180,"publishedAt":44},"01M6DZRBD8G9E6AGM3VA87VRZG","https:\u002F\u002Fgithub.com\u002Fconstellation-works\u002Forbit\u002Fblob\u002Fagent-main\u002Fcrates\u002Forbit-core\u002Fassets\u002Fskills\u002Forbit-orchestrate\u002FSKILL.md",[179],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[26,24,25,43,42,41,40,39,38,37,36,35,34,33,32,31,30,29,28,27]]