[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6WWQ2V1SW5CBTK11NERGR5G":3,"skill-same:01M6WWQ2V1SW5CBTK11NERGR5G":107},{"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":41,"linkOnly":42,"markdownContent":43,"markdownBody":44,"frontMatter":45,"refName":59,"commitSha":60,"skillDirectory":61,"forksCount":14,"repositoryHtmlUrl":62,"repositoryHomepageUrl":63,"ownerHtmlUrl":64,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":65,"install":66,"resources":78},"01M6WWQ2V1SW5CBTK11NERGR5G","task-router",null,"Route a task description to the right processkit skill, project-specific process override, and MCP tool in a single deterministic call — use this at the start of every processkit domain task.","projectious-work\u002Fprocesskit","projectious-work","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F263889726?v=4","01M6N97TJJ118WQ35CHRV8RD0E","projectious.work","01M6N97TPWJPF0BMKYQ4PBY5W8",0,"MIT","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Farchive\u002Fprocesskit-install-20261101\u002Fcontext\u002Fskills\u002Fprocesskit\u002Ftask-router\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],"task router","router","task","call","single","at","true","project","domain","deterministic","route","MCP","use","process","work","tool","start","2026-11-10 10:12:30",false,"---\nname: task-router\ndescription: >\n  Route a task description to the right processkit skill, project-specific\n  process override, and MCP tool in a single deterministic call — use this\n  at the start of every processkit domain task.\nmetadata:\n  processkit:\n    apiVersion: processkit.projectious.work\u002Fv2\n    id: SKILL-task-router\n    version: \"1.0.0\"\n    created: 2026-04-11T00:00:00Z\n    category: processkit\n    layer: 1\n    core: true\n    provides:\n      mcp_tools:\n        - route_task\n---\n\n# Task Router\n\n## Intro\n\n`task-router` is the primary routing entry point for processkit agents.\nCall `route_task(task_description)` once at the start of any domain\ntask. It returns — in a single call, without an LLM — the matching\nskill, the legacy project-specific process override (if any), and the\nrecommended MCP tool. Use it instead of calling `find_skill()` directly.\n\n## Overview\n\n### When to use\n\nCall `route_task()` before any `create_*`, `transition_*`, `link_*`,\nor `record_*` MCP tool. The 1% rule applies: if there is even a 1%\nchance a processkit skill covers this task, route first.\n\n### Routing algorithm\n\nTwo-phase heuristic, no LLM call required:\n\n1. **Phase 1 — domain group match** (cheap keyword classifier).\n   Scores the task against 13 domain groups by keyword overlap.\n   Eliminates ~90% of candidates in one pass.\n\n2. **Phase 2 — tool ranking** (token overlap or local embedding-style\n   cosine scoring within the matched group). Scores the ~3–6 tools in\n   the matched group. Returns the top tool with rationale and two\n   runner-up candidates.\n\n3. **Fallback** — if no group scores ≥ 0.3, falls back to the\n   skill-finder trigger-phrase table for cross-domain or meta tasks.\n\n### Reading the result\n\n| Field | What it means |\n|---|---|\n| `skill` | Load this skill's SKILL.md before acting |\n| `skill_description_excerpt` | First 150 chars of skill description — judge fit without loading the full file |\n| `process_override` | Legacy v1 compatibility. If present, read this file before the skill; it contains project-specific overrides from `context\u002Fprocesses\u002F` |\n| `process_override_status` | Present with `process_override`; currently `legacy-v1` |\n| `tool_qualified` | `{server}__{tool}` — collision-safe tool identifier |\n| `confidence` | 0.0–1.0 geometric mean of group and tool scores |\n| `routing_basis` | `keyword_match` = confident; `needs_llm_confirm` = escalate to user\u002FLLM |\n| `candidate_tools[]` | Top-3 scored tools with rationales for fallback |\n\n### When `confidence \u003C 0.5`\n\n`routing_basis` will be `needs_llm_confirm`. Show `candidate_tools[]`\nto the user or pass them to an LLM for disambiguation. Do NOT call a\n`create_*` tool without confirmation when confidence is low.\n\nCallers may pass `allow_llm_escalation=true`; the router then returns a\n`llm_escalation` status with the configured `fast` model class hint. The\nstdio MCP server does not make provider network calls itself, so the caller\nowns the actual cheap-model confirmation step.\n\n### When `process_override` is present\n\nRead the process override file **before** the generic skill. It\ncontains project-specific step overrides, gates, and blockers that\nsupersede the skill defaults.\n\nThis is a legacy v1 compatibility surface. v2 processkit releases do not\nship first-class Process entities or `src\u002Fcontext\u002Fprocesses\u002F`. Existing\nprojects may still carry live `context\u002Fprocesses\u002F*.md` files as migration\nsource state; when found, the router returns them with\n`process_override_status: legacy-v1`.\n\n## Gotchas\n\n- **Do not call `find_skill()` directly** in normal flow — `route_task()`\n  calls it internally as a fallback. Direct `find_skill()` calls skip\n  the process override lookup and the domain group fast path.\n- **Low confidence is a signal, not a failure.** Confidence \u003C 0.5 means\n  the task straddles multiple domains or uses uncommon vocabulary. Use\n  `candidate_tools[]` to present options.\n- **Task description quality matters.** Verb-noun phrases score best:\n  \"create work item\", \"record decision\", \"log event\". Single words\n  (\"task\", \"log\") are ambiguous — add context.\n- **Process overrides are compatibility data.** Do not create new v2\n  Process entities for router overrides. Prefer skill configuration,\n  definition Artifacts, or WorkItems that describe process instances.\n\n## Full reference\n\n### MCP server\n\nSee `mcp\u002FSERVER.md` for tool schema and return-shape details.\n\n### Domain groups\n\nThe 13 registered groups: `workitem`, `decision`, `discussion`,\n`event`, `actor`, `scope`, `index`, `skill`, `gate`, `model`,\n`id`, `role`, `binding`.\n\n### Extending the router\n\nTo add a new domain group:\n1. Add an entry to `DOMAIN_GROUPS` in `mcp\u002Fserver.py`.\n2. Add trigger phrases to the `skill-finder` trigger table.\n3. Run the smoke tests: `uv run scripts\u002Fsmoke-test-servers.py`.\n","# Task Router\n\n## Intro\n\n`task-router` is the primary routing entry point for processkit agents.\nCall `route_task(task_description)` once at the start of any domain\ntask. It returns — in a single call, without an LLM — the matching\nskill, the legacy project-specific process override (if any), and the\nrecommended MCP tool. Use it instead of calling `find_skill()` directly.\n\n## Overview\n\n### When to use\n\nCall `route_task()` before any `create_*`, `transition_*`, `link_*`,\nor `record_*` MCP tool. The 1% rule applies: if there is even a 1%\nchance a processkit skill covers this task, route first.\n\n### Routing algorithm\n\nTwo-phase heuristic, no LLM call required:\n\n1. **Phase 1 — domain group match** (cheap keyword classifier).\n   Scores the task against 13 domain groups by keyword overlap.\n   Eliminates ~90% of candidates in one pass.\n\n2. **Phase 2 — tool ranking** (token overlap or local embedding-style\n   cosine scoring within the matched group). Scores the ~3–6 tools in\n   the matched group. Returns the top tool with rationale and two\n   runner-up candidates.\n\n3. **Fallback** — if no group scores ≥ 0.3, falls back to the\n   skill-finder trigger-phrase table for cross-domain or meta tasks.\n\n### Reading the result\n\n| Field | What it means |\n|---|---|\n| `skill` | Load this skill's SKILL.md before acting |\n| `skill_description_excerpt` | First 150 chars of skill description — judge fit without loading the full file |\n| `process_override` | Legacy v1 compatibility. If present, read this file before the skill; it contains project-specific overrides from `context\u002Fprocesses\u002F` |\n| `process_override_status` | Present with `process_override`; currently `legacy-v1` |\n| `tool_qualified` | `{server}__{tool}` — collision-safe tool identifier |\n| `confidence` | 0.0–1.0 geometric mean of group and tool scores |\n| `routing_basis` | `keyword_match` = confident; `needs_llm_confirm` = escalate to user\u002FLLM |\n| `candidate_tools[]` | Top-3 scored tools with rationales for fallback |\n\n### When `confidence \u003C 0.5`\n\n`routing_basis` will be `needs_llm_confirm`. Show `candidate_tools[]`\nto the user or pass them to an LLM for disambiguation. Do NOT call a\n`create_*` tool without confirmation when confidence is low.\n\nCallers may pass `allow_llm_escalation=true`; the router then returns a\n`llm_escalation` status with the configured `fast` model class hint. The\nstdio MCP server does not make provider network calls itself, so the caller\nowns the actual cheap-model confirmation step.\n\n### When `process_override` is present\n\nRead the process override file **before** the generic skill. It\ncontains project-specific step overrides, gates, and blockers that\nsupersede the skill defaults.\n\nThis is a legacy v1 compatibility surface. v2 processkit releases do not\nship first-class Process entities or `src\u002Fcontext\u002Fprocesses\u002F`. Existing\nprojects may still carry live `context\u002Fprocesses\u002F*.md` files as migration\nsource state; when found, the router returns them with\n`process_override_status: legacy-v1`.\n\n## Gotchas\n\n- **Do not call `find_skill()` directly** in normal flow — `route_task()`\n  calls it internally as a fallback. Direct `find_skill()` calls skip\n  the process override lookup and the domain group fast path.\n- **Low confidence is a signal, not a failure.** Confidence \u003C 0.5 means\n  the task straddles multiple domains or uses uncommon vocabulary. Use\n  `candidate_tools[]` to present options.\n- **Task description quality matters.** Verb-noun phrases score best:\n  \"create work item\", \"record decision\", \"log event\". Single words\n  (\"task\", \"log\") are ambiguous — add context.\n- **Process overrides are compatibility data.** Do not create new v2\n  Process entities for router overrides. Prefer skill configuration,\n  definition Artifacts, or WorkItems that describe process instances.\n\n## Full reference\n\n### MCP server\n\nSee `mcp\u002FSERVER.md` for tool schema and return-shape details.\n\n### Domain groups\n\nThe 13 registered groups: `workitem`, `decision`, `discussion`,\n`event`, `actor`, `scope`, `index`, `skill`, `gate`, `model`,\n`id`, `role`, `binding`.\n\n### Extending the router\n\nTo add a new domain group:\n1. Add an entry to `DOMAIN_GROUPS` in `mcp\u002Fserver.py`.\n2. Add trigger phrases to the `skill-finder` trigger table.\n3. Run the smoke tests: `uv run scripts\u002Fsmoke-test-servers.py`.\n",{"name":5,"metadata":46,"description":58},{"processkit":47},{"id":48,"core":49,"layer":50,"created":51,"version":52,"category":53,"provides":54,"apiVersion":57},"SKILL-task-router",true,1,1775865600000,"1.0.0","processkit",{"mcp_tools":55},[56],"route_task","processkit.projectious.work\u002Fv2","Route a task description to the right processkit skill, project-specific process override, and MCP tool in a single deterministic call — use this at the start of every processkit domain task.\n","main","4876a92620ee792aac2ac9a1a806322d8a572367","archive\u002Fprocesskit-install-20261101\u002Fcontext\u002Fskills\u002Fprocesskit\u002Ftask-router","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit","https:\u002F\u002Fprojectious-work.github.io\u002Fprocesskit\u002F","https:\u002F\u002Fgithub.com\u002Fprojectious-work","2026-11-11 23:24:18",{"source":8,"skillName":5,"commands":67,"prompt":77,"zipAvailable":49},[68,71,74],{"tool":69,"command":70},"npx","npx skills add projectious-work\u002Fprocesskit --skill task-router",{"tool":72,"command":73},"bunx","bunx skills add projectious-work\u002Fprocesskit --skill task-router",{"tool":75,"command":76},"pnpm","pnpm dlx skills add projectious-work\u002Fprocesskit --skill task-router","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Farchive\u002Fprocesskit-install-20261101\u002Fcontext\u002Fskills\u002Fprocesskit\u002Ftask-router\u002FSKILL.md",[79,85,89,95,99,103],{"path":80,"resourceType":81,"resourceTypeDesc":82,"sizeBytes":83,"sourceHtmlUrl":84},"mcp\u002Fmcp-config.json","REFERENCE","参考文档",205,"https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Farchive\u002Fprocesskit-install-20261101\u002Fcontext\u002Fskills\u002Fprocesskit\u002Ftask-router\u002Fmcp\u002Fmcp-config.json",{"path":86,"resourceType":81,"resourceTypeDesc":82,"sizeBytes":87,"sourceHtmlUrl":88},"mcp\u002FSERVER.md",2486,"https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Farchive\u002Fprocesskit-install-20261101\u002Fcontext\u002Fskills\u002Fprocesskit\u002Ftask-router\u002Fmcp\u002FSERVER.md",{"path":90,"resourceType":91,"resourceTypeDesc":92,"sizeBytes":93,"sourceHtmlUrl":94},"mcp\u002Fserver.py","SCRIPT","脚本",51819,"https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Farchive\u002Fprocesskit-install-20261101\u002Fcontext\u002Fskills\u002Fprocesskit\u002Ftask-router\u002Fmcp\u002Fserver.py",{"path":96,"resourceType":81,"resourceTypeDesc":82,"sizeBytes":97,"sourceHtmlUrl":98},"mcp\u002Fuser_config.json",406,"https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Farchive\u002Fprocesskit-install-20261101\u002Fcontext\u002Fskills\u002Fprocesskit\u002Ftask-router\u002Fmcp\u002Fuser_config.json",{"path":100,"resourceType":91,"resourceTypeDesc":92,"sizeBytes":101,"sourceHtmlUrl":102},"scripts\u002Ftest_task_router.py",5361,"https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Farchive\u002Fprocesskit-install-20261101\u002Fcontext\u002Fskills\u002Fprocesskit\u002Ftask-router\u002Fscripts\u002Ftest_task_router.py",{"path":104,"resourceType":91,"resourceTypeDesc":92,"sizeBytes":105,"sourceHtmlUrl":106},"tests\u002Ftest_engineering_role_coverage.py",7233,"https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Farchive\u002Fprocesskit-install-20261101\u002Fcontext\u002Fskills\u002Fprocesskit\u002Ftask-router\u002Ftests\u002Ftest_engineering_role_coverage.py",[108,115,122,128,151,169],{"skillNo":109,"name":110,"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":111,"categories":112,"keywords":113,"publishedAt":114},"01M7665G2ED3624R3X93304265","standard-ui-design","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fstandard-ui-design\u002FSKILL.md",[],[],"2026-11-12 13:55:43",{"skillNo":116,"name":117,"nameZh":6,"description":118,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":119,"categories":120,"keywords":121,"publishedAt":114},"01M76651QTC6W0HJ4AFW2MMR3V","standard-agent-use","Company standard for working with AI agents. Use when deciding what to delegate to agents, capturing open work, keeping agent memory or knowledge notes, improving an agent workflow, or introducing agent work in a team or for a client.","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fstandard-agent-use\u002FSKILL.md",[],[],{"skillNo":123,"name":124,"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":125,"categories":126,"keywords":127,"publishedAt":114},"01M76650NN4T8P241021NBR64S","standard-agent-building","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fstandard-agent-building\u002FSKILL.md",[],[],{"skillNo":129,"name":130,"nameZh":6,"description":131,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":132,"categories":133,"keywords":135,"publishedAt":41},"01M6X5MTC89B4ZGXW3JVSQCPF8","skill-management","Manage Skill package manifests as v1 ontology entities.","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Fsrc\u002Fcontext\u002Fskills\u002Fprocesskit\u002Fskill-management\u002FSKILL.md",[134],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[136,137,138,139,140,141,38,142,143,144,145,146,130,147,148,149,150],"management","skill management","ontology","v1","create","as","transition","get","manage","update","alpha","list","package","skill package","manifests",{"skillNo":152,"name":153,"nameZh":6,"description":154,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":155,"categories":156,"keywords":158,"publishedAt":41},"01M6X5MCV74NHW5MEDNHZRF2C8","schema-management","Manage generated processkit schema contracts through deterministic MCP rebuild and inspection workflows.","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Fsrc\u002Fcontext\u002Fskills\u002Fprocesskit\u002Fschema-management\u002FSKILL.md",[157],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[159,160,136,161,162,163,164,165,166,35,167,33,38,143,144,146,168],"schema","schema management","contract","rebuild","contracts","workflows","wed","generated","validation","mode",{"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":41},"01M6X5M5G6YJ96AMZGSCZJ1D13","proposition-management","Record claims, risks, beliefs, world facts, WSJF estimates, and assumptions through the Proposition interface.","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Fsrc\u002Fcontext\u002Fskills\u002Fprocesskit\u002Fproposition-management\u002FSKILL.md",[],[176,136,177,178,140,179,180,181,38,143,182,183,145,184,146,185,186,187,188,189],"proposition management","proposition","interface","claims","record","world","creation","event","WSJF","index","updates","query","facts","log"]