[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6NQJFD1NHHGRK2GKRR389JK":3,"skill-same:01M6NQJFD1NHHGRK2GKRR389JK":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":69,"commitSha":70,"skillDirectory":71,"forksCount":14,"repositoryHtmlUrl":72,"repositoryHomepageUrl":73,"ownerHtmlUrl":74,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":75,"install":76,"resources":89},"01M6NQJFD1NHHGRK2GKRR389JK","runtime-prune",null,"Inspect, plan, and invoke safe cleanup for runtime-manager-owned state without binding the workflow to a specific host orchestrator or agent provider.","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\u002Fsrc\u002Fcontext\u002Fskills\u002Fprocesskit\u002Fruntime-prune\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],"runtime prune","prune","runtime","orchestrator","invoke","apply","for","inspect","usage","host","provider","workflow","confirmation","work","analyze","plan","disk","state","routing","cleanup","2026-11-10 10:12:30",false,"---\nname: runtime-prune\ndescription: >\n  Inspect, plan, and invoke safe cleanup for runtime-manager-owned state\n  without binding the workflow to a specific host orchestrator or agent\n  provider.\nmetadata:\n  processkit:\n    apiVersion: processkit.projectious.work\u002Fv2\n    id: SKILL-runtime-prune\n    version: \"1.0.0\"\n    created: 2026-05-13T00:00:00Z\n    category: processkit\n    layer: 2\n    uses:\n      - skill: skill-gate\n        purpose: >\n          Require explicit routing and confirmation before destructive\n          cleanup.\n    provides:\n      primitives: []\n      mcp_tools:\n        - analyze_disk_usage\n        - plan_prune\n        - apply_prune\n      assets: []\n---\n\n# Runtime Prune\n\n## Intro\n\n`runtime-prune` gives agents a provider-neutral cleanup workflow for\nruntime-manager-owned state, repo-local generated artifacts, and remote\nquota surfaces. It inventories selected cleanup scopes, produces a\ndry-run plan, applies low\u002Fmedium-risk processkit-owned cleanup from an\nexplicit allowlist, and returns structured host-action evidence for\nhigh-risk runtime-owner or provider cleanup. It does not invoke host\norchestrator or provider deletion commands from inside the processkit\nruntime.\n\n## Overview\n\nUse this skill when a project has accumulated runtime-home caches,\nbuild caches, repo-local release\u002Fbuild artifacts, package-manager\ncaches, agent worktree state, container storage, E2E companion state,\nor remote forge storage such as CI\u002Faction artifacts, release assets, or\npackage\u002Fcontainer registry versions.\n\nThe workflow is:\n\n1. Call `analyze_disk_usage(project_root, scopes)` to inventory\n   configured cleanup scopes without modifying the filesystem.\n2. Call `plan_prune(project_root, scopes)` to produce a dry-run plan,\n   expected reclaimed bytes, risk labels, and the exact confirmation\n   token needed for apply.\n3. Call `apply_prune(project_root, scopes, confirmation)` only after\n   the user has explicitly approved the plan. The apply tool may remove\n   only explicit processkit-owned allowlist targets; provider\u002Fruntime\n   owner cleanup returns an external host-action result for the owner.\n4. For external provider scopes, use each action's `inventory_command`\n   and `dry_run_command` first. The `apply_command` is concrete enough\n   to run after human approval when the derived-project agent has the\n   required CLI, token, and environment variables.\n\n### MCP Tools\n\n| Tool | Purpose |\n|---|---|\n| `analyze_disk_usage(project_root, scopes)` | Return structured size and risk data for selected cleanup scopes. |\n| `plan_prune(project_root, scopes)` | Return a dry-run prune plan and required confirmation token. |\n| `apply_prune(project_root, scopes, confirmation)` | Apply processkit allowlist cleanup and return external host-action evidence for unsupported scopes. |\n\n### Scopes\n\n| Scope | What it covers | Apply policy |\n|---|---|---|\n| `runtime-home` | Bounded runtime-home cache, diagnostics, tmp, and runtime paths. | Processkit allowlist cleanup. |\n| `build-cache` | Explicit build-cache paths such as Rust incremental and build subdirectories. | Processkit allowlist cleanup. |\n| `repo-artifacts` | Generated local artifacts such as `dist\u002F`, `build\u002F`, coverage, and test caches. | Processkit allowlist cleanup. |\n| `tool-caches` | Project-local package\u002Ftool caches such as pip, uv, npm, pnpm, and node build caches. | Processkit allowlist cleanup. |\n| `agent-worktrees` | Provider-created or runtime-manager-owned Git worktrees discovered from Git metadata. | External host action after explicit confirmation. |\n| `containers` | Host runtime container cleanup. | External host action; no direct Docker or Podman calls. |\n| `e2e-companion` | Nested companion environment cleanup. | External host action; companion access must use owner-approved reachability checks. |\n| `action-artifacts` | Remote CI\u002Faction artifacts, logs, and stale workflow artifacts. | External provider action; no direct deletion. |\n| `release-assets` | Remote release binary assets older than the support\u002Fretention window. | External provider action; no direct deletion. |\n| `package-registry` | Remote package\u002Fcontainer registry versions. | External provider action; no direct deletion. |\n\n### External Runbooks\n\nFor external scopes, `plan_prune()` returns a provider-adapted runbook\ninstead of deleting anything:\n\n- `required_env`: variables the derived-project agent or human must set.\n- `inventory_command`: read-only command that lists current artifacts.\n- `dry_run_command`: read-only command that prints deletion candidates\n  and, when the provider exposes sizes, a byte total.\n- `apply_command`: deletion command using the same candidate logic as\n  the dry run. Run it only after explicit human approval.\n- `space_estimate`: how to interpret `x` and `y` for \"delete x and gain\n  y\" advice when the provider does or does not expose byte totals.\n\nProvider adapters currently cover GitHub, GitLab, Codeberg, Forgejo,\nGitea, and generic\u002Funknown forges. Unknown providers intentionally\nreturn empty commands plus a note to fill the provider API details.\n\n## Gotchas\n\n- **Do not treat analysis as approval.** `analyze_disk_usage()` and\n  `plan_prune()` are read-only. Apply still needs the exact confirmation\n  token returned by the plan.\n- **Do not hand-roll destructive provider cleanup.** `apply_prune()` may\n  clean processkit-owned allowlist targets, but worktrees, containers,\n  and companion state must return structured external host-action\n  evidence instead of invoking a host orchestrator command.\n- **Dry-run external cleanup first.** For `action-artifacts`,\n  `release-assets`, and `package-registry`, use `inventory_command` and\n  `dry_run_command` before proposing `apply_command`.\n- **Do not remove the whole runtime home.** `runtime-home` targets only\n  narrow cache, diagnostics, tmp, and runtime subpaths, not the whole\n  `.aibox-home` tree.\n- **Do not inspect local Docker or Podman for E2E companion cleanup.**\n  Companion state is remote from the devcontainer's perspective; use\n  the owner-approved host workflow that checks reachability.\n- **Do not count arbitrary project files as reclaimable.** Only\n  explicitly scoped runtime and cache paths contribute to expected\n  reclaimed bytes.\n- **Do not delete remote packages by age alone.** Keep source tags,\n  release metadata, active deployment tags, and the supported-version\n  window; delete binary assets or package versions only after an owner\n  approves the exact target list.\n- **Do not use Docker big-hammer cleanup by default.** Inspect first,\n  prune builder cache separately, and avoid volume deletion unless the\n  volume data owner is explicit.\n- **Do not delete provider worktrees by path heuristics.** Worktrees\n  are inventoried for visibility, but deletion is delegated to the host\n  runtime manager because it owns provider-specific safety checks.\n- **Keep paths provider-neutral in user-facing output.** Report scopes\n  like `agent-worktrees` and `runtime-home`, not Claude-, Codex-, or\n  shell-specific state unless a discovered path itself contains that\n  name.\n\n## Best-practice briefing\n\n- Inventory before deletion: local `du`\u002Ftool reports, Docker\n  `system df`\u002FBuildKit usage, and repository-provider API listings\n  should produce the target list before any apply step.\n- Separate cleanup classes: generated repo artifacts, dependency\n  caches, Docker build cache, Docker images\u002Fvolumes, Actions artifacts,\n  release assets, and registry package versions have different owners\n  and risk profiles.\n- Prefer retention policy over emergency cleanup: set short\n  `retention-days` for non-release Actions artifacts, keep only the\n  supported release-asset window, and apply registry lifecycle rules or\n  scheduled cleanup for untagged\u002Fsuperseded images.\n- Preserve reproducibility evidence: tags, release notes, checksums,\n  SBOMs, provenance, and currently supported binaries should survive\n  cleanup unless a release policy says otherwise.\n- Use provider-neutral scopes first. `runtime-prune` adapts host-action\n  hints for GitHub, GitLab, Codeberg, Forgejo, Gitea, and generic\n  providers based on the repository remote.\n\n## Full reference\n\n`analyze_disk_usage()` resolves `project_root` to an existing directory\nand defaults to all scopes when none are supplied. It never follows\nsymlinks during size walks.\n\n`plan_prune()` returns:\n\n- `required_confirmation`, formatted as\n  `apply-prune:\u003Ccomma-separated-scopes>`\n- per-scope actions with `risk`, `apply_owner`,\n  `expected_reclaim_bytes`, `targets`, `required_env`,\n  `inventory_command`, `dry_run_command`, `apply_command`, and\n  `space_estimate`\n\n`apply_prune()` accepts only the exact `required_confirmation` value\nfrom the matching plan. For `runtime-home` and `build-cache`, it removes\nonly the allowlisted paths from the plan. For external scopes, it records\nper-scope host-action evidence and leaves execution to the owner outside\nthe container.\n","# Runtime Prune\n\n## Intro\n\n`runtime-prune` gives agents a provider-neutral cleanup workflow for\nruntime-manager-owned state, repo-local generated artifacts, and remote\nquota surfaces. It inventories selected cleanup scopes, produces a\ndry-run plan, applies low\u002Fmedium-risk processkit-owned cleanup from an\nexplicit allowlist, and returns structured host-action evidence for\nhigh-risk runtime-owner or provider cleanup. It does not invoke host\norchestrator or provider deletion commands from inside the processkit\nruntime.\n\n## Overview\n\nUse this skill when a project has accumulated runtime-home caches,\nbuild caches, repo-local release\u002Fbuild artifacts, package-manager\ncaches, agent worktree state, container storage, E2E companion state,\nor remote forge storage such as CI\u002Faction artifacts, release assets, or\npackage\u002Fcontainer registry versions.\n\nThe workflow is:\n\n1. Call `analyze_disk_usage(project_root, scopes)` to inventory\n   configured cleanup scopes without modifying the filesystem.\n2. Call `plan_prune(project_root, scopes)` to produce a dry-run plan,\n   expected reclaimed bytes, risk labels, and the exact confirmation\n   token needed for apply.\n3. Call `apply_prune(project_root, scopes, confirmation)` only after\n   the user has explicitly approved the plan. The apply tool may remove\n   only explicit processkit-owned allowlist targets; provider\u002Fruntime\n   owner cleanup returns an external host-action result for the owner.\n4. For external provider scopes, use each action's `inventory_command`\n   and `dry_run_command` first. The `apply_command` is concrete enough\n   to run after human approval when the derived-project agent has the\n   required CLI, token, and environment variables.\n\n### MCP Tools\n\n| Tool | Purpose |\n|---|---|\n| `analyze_disk_usage(project_root, scopes)` | Return structured size and risk data for selected cleanup scopes. |\n| `plan_prune(project_root, scopes)` | Return a dry-run prune plan and required confirmation token. |\n| `apply_prune(project_root, scopes, confirmation)` | Apply processkit allowlist cleanup and return external host-action evidence for unsupported scopes. |\n\n### Scopes\n\n| Scope | What it covers | Apply policy |\n|---|---|---|\n| `runtime-home` | Bounded runtime-home cache, diagnostics, tmp, and runtime paths. | Processkit allowlist cleanup. |\n| `build-cache` | Explicit build-cache paths such as Rust incremental and build subdirectories. | Processkit allowlist cleanup. |\n| `repo-artifacts` | Generated local artifacts such as `dist\u002F`, `build\u002F`, coverage, and test caches. | Processkit allowlist cleanup. |\n| `tool-caches` | Project-local package\u002Ftool caches such as pip, uv, npm, pnpm, and node build caches. | Processkit allowlist cleanup. |\n| `agent-worktrees` | Provider-created or runtime-manager-owned Git worktrees discovered from Git metadata. | External host action after explicit confirmation. |\n| `containers` | Host runtime container cleanup. | External host action; no direct Docker or Podman calls. |\n| `e2e-companion` | Nested companion environment cleanup. | External host action; companion access must use owner-approved reachability checks. |\n| `action-artifacts` | Remote CI\u002Faction artifacts, logs, and stale workflow artifacts. | External provider action; no direct deletion. |\n| `release-assets` | Remote release binary assets older than the support\u002Fretention window. | External provider action; no direct deletion. |\n| `package-registry` | Remote package\u002Fcontainer registry versions. | External provider action; no direct deletion. |\n\n### External Runbooks\n\nFor external scopes, `plan_prune()` returns a provider-adapted runbook\ninstead of deleting anything:\n\n- `required_env`: variables the derived-project agent or human must set.\n- `inventory_command`: read-only command that lists current artifacts.\n- `dry_run_command`: read-only command that prints deletion candidates\n  and, when the provider exposes sizes, a byte total.\n- `apply_command`: deletion command using the same candidate logic as\n  the dry run. Run it only after explicit human approval.\n- `space_estimate`: how to interpret `x` and `y` for \"delete x and gain\n  y\" advice when the provider does or does not expose byte totals.\n\nProvider adapters currently cover GitHub, GitLab, Codeberg, Forgejo,\nGitea, and generic\u002Funknown forges. Unknown providers intentionally\nreturn empty commands plus a note to fill the provider API details.\n\n## Gotchas\n\n- **Do not treat analysis as approval.** `analyze_disk_usage()` and\n  `plan_prune()` are read-only. Apply still needs the exact confirmation\n  token returned by the plan.\n- **Do not hand-roll destructive provider cleanup.** `apply_prune()` may\n  clean processkit-owned allowlist targets, but worktrees, containers,\n  and companion state must return structured external host-action\n  evidence instead of invoking a host orchestrator command.\n- **Dry-run external cleanup first.** For `action-artifacts`,\n  `release-assets`, and `package-registry`, use `inventory_command` and\n  `dry_run_command` before proposing `apply_command`.\n- **Do not remove the whole runtime home.** `runtime-home` targets only\n  narrow cache, diagnostics, tmp, and runtime subpaths, not the whole\n  `.aibox-home` tree.\n- **Do not inspect local Docker or Podman for E2E companion cleanup.**\n  Companion state is remote from the devcontainer's perspective; use\n  the owner-approved host workflow that checks reachability.\n- **Do not count arbitrary project files as reclaimable.** Only\n  explicitly scoped runtime and cache paths contribute to expected\n  reclaimed bytes.\n- **Do not delete remote packages by age alone.** Keep source tags,\n  release metadata, active deployment tags, and the supported-version\n  window; delete binary assets or package versions only after an owner\n  approves the exact target list.\n- **Do not use Docker big-hammer cleanup by default.** Inspect first,\n  prune builder cache separately, and avoid volume deletion unless the\n  volume data owner is explicit.\n- **Do not delete provider worktrees by path heuristics.** Worktrees\n  are inventoried for visibility, but deletion is delegated to the host\n  runtime manager because it owns provider-specific safety checks.\n- **Keep paths provider-neutral in user-facing output.** Report scopes\n  like `agent-worktrees` and `runtime-home`, not Claude-, Codex-, or\n  shell-specific state unless a discovered path itself contains that\n  name.\n\n## Best-practice briefing\n\n- Inventory before deletion: local `du`\u002Ftool reports, Docker\n  `system df`\u002FBuildKit usage, and repository-provider API listings\n  should produce the target list before any apply step.\n- Separate cleanup classes: generated repo artifacts, dependency\n  caches, Docker build cache, Docker images\u002Fvolumes, Actions artifacts,\n  release assets, and registry package versions have different owners\n  and risk profiles.\n- Prefer retention policy over emergency cleanup: set short\n  `retention-days` for non-release Actions artifacts, keep only the\n  supported release-asset window, and apply registry lifecycle rules or\n  scheduled cleanup for untagged\u002Fsuperseded images.\n- Preserve reproducibility evidence: tags, release notes, checksums,\n  SBOMs, provenance, and currently supported binaries should survive\n  cleanup unless a release policy says otherwise.\n- Use provider-neutral scopes first. `runtime-prune` adapts host-action\n  hints for GitHub, GitLab, Codeberg, Forgejo, Gitea, and generic\n  providers based on the repository remote.\n\n## Full reference\n\n`analyze_disk_usage()` resolves `project_root` to an existing directory\nand defaults to all scopes when none are supplied. It never follows\nsymlinks during size walks.\n\n`plan_prune()` returns:\n\n- `required_confirmation`, formatted as\n  `apply-prune:\u003Ccomma-separated-scopes>`\n- per-scope actions with `risk`, `apply_owner`,\n  `expected_reclaim_bytes`, `targets`, `required_env`,\n  `inventory_command`, `dry_run_command`, `apply_command`, and\n  `space_estimate`\n\n`apply_prune()` accepts only the exact `required_confirmation` value\nfrom the matching plan. For `runtime-home` and `build-cache`, it removes\nonly the allowlisted paths from the plan. For external scopes, it records\nper-scope host-action evidence and leaves execution to the owner outside\nthe container.\n",{"name":5,"metadata":49,"description":68},{"processkit":50},{"id":51,"uses":52,"layer":56,"created":57,"version":58,"category":59,"provides":60,"apiVersion":67},"SKILL-runtime-prune",[53],{"skill":54,"purpose":55},"skill-gate","Require explicit routing and confirmation before destructive cleanup.\n",2,1778630400000,"1.0.0","processkit",{"assets":61,"mcp_tools":62,"primitives":66},[],[63,64,65],"analyze_disk_usage","plan_prune","apply_prune",[],"processkit.projectious.work\u002Fv2","Inspect, plan, and invoke safe cleanup for runtime-manager-owned state without binding the workflow to a specific host orchestrator or agent provider.\n","main","7231f100e266e30d2c1699ecde5c00602471dc3c","src\u002Fcontext\u002Fskills\u002Fprocesskit\u002Fruntime-prune","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":77,"prompt":87,"zipAvailable":88},[78,81,84],{"tool":79,"command":80},"npx","npx skills add projectious-work\u002Fprocesskit --skill runtime-prune",{"tool":82,"command":83},"bunx","bunx skills add projectious-work\u002Fprocesskit --skill runtime-prune",{"tool":85,"command":86},"pnpm","pnpm dlx skills add projectious-work\u002Fprocesskit --skill runtime-prune","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Fsrc\u002Fcontext\u002Fskills\u002Fprocesskit\u002Fruntime-prune\u002FSKILL.md",true,[90,96,100,106],{"path":91,"resourceType":92,"resourceTypeDesc":93,"sizeBytes":94,"sourceHtmlUrl":95},"mcp\u002Fmcp-config.json","REFERENCE","参考文档",209,"https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Fsrc\u002Fcontext\u002Fskills\u002Fprocesskit\u002Fruntime-prune\u002Fmcp\u002Fmcp-config.json",{"path":97,"resourceType":92,"resourceTypeDesc":93,"sizeBytes":98,"sourceHtmlUrl":99},"mcp\u002FSERVER.md",1514,"https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Fsrc\u002Fcontext\u002Fskills\u002Fprocesskit\u002Fruntime-prune\u002Fmcp\u002FSERVER.md",{"path":101,"resourceType":102,"resourceTypeDesc":103,"sizeBytes":104,"sourceHtmlUrl":105},"mcp\u002Fserver.py","SCRIPT","脚本",39961,"https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Fsrc\u002Fcontext\u002Fskills\u002Fprocesskit\u002Fruntime-prune\u002Fmcp\u002Fserver.py",{"path":107,"resourceType":102,"resourceTypeDesc":103,"sizeBytes":108,"sourceHtmlUrl":109},"scripts\u002Ftest_runtime_prune.py",9683,"https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Fsrc\u002Fcontext\u002Fskills\u002Fprocesskit\u002Fruntime-prune\u002Fscripts\u002Ftest_runtime_prune.py",[111,118,125,131,154,174],{"skillNo":112,"name":113,"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":114,"categories":115,"keywords":116,"publishedAt":117},"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":119,"name":120,"nameZh":6,"description":121,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":122,"categories":123,"keywords":124,"publishedAt":117},"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":126,"name":127,"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":128,"categories":129,"keywords":130,"publishedAt":117},"01M76650NN4T8P241021NBR64S","standard-agent-building","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fstandard-agent-building\u002FSKILL.md",[],[],{"skillNo":132,"name":133,"nameZh":6,"description":134,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":135,"categories":136,"keywords":138,"publishedAt":44},"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",[137],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[139,140,141,142,143,144,37,145,146,147,148,149,133,150,151,152,153],"management","skill management","ontology","v1","create","as","transition","get","manage","update","alpha","list","package","skill package","manifests",{"skillNo":155,"name":156,"nameZh":6,"description":157,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":158,"categories":159,"keywords":161,"publishedAt":44},"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",[160],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[162,163,139,164,165,166,167,168,169,170,171,172,37,146,147,149,173],"schema","schema management","contract","rebuild","contracts","workflows","wed","generated","MCP","validation","deterministic","mode",{"skillNo":175,"name":176,"nameZh":6,"description":177,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":178,"categories":179,"keywords":180,"publishedAt":44},"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",[],[181,139,182,183,143,184,185,186,37,146,187,188,148,189,149,190,191,192,193,194],"proposition management","proposition","interface","claims","record","world","creation","event","WSJF","index","updates","query","facts","log"]