[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6X16ESPZ9DN8K5RZMCNSW5G":3,"skill-same:01M6X16ESPZ9DN8K5RZMCNSW5G":113},{"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":26,"publishedAt":47,"linkOnly":48,"markdownContent":49,"markdownBody":50,"frontMatter":51,"refName":70,"commitSha":71,"skillDirectory":72,"forksCount":14,"repositoryHtmlUrl":73,"repositoryHomepageUrl":74,"ownerHtmlUrl":75,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":76,"install":77,"resources":90},"01M6X16ESPZ9DN8K5RZMCNSW5G","index-management",null,"SQLite-backed index over all entity files in the project. The read-side foundation for every other MCP server. Use whenever an agent needs to look up entities by ID, kind, state, or text — instead of grepping the filesystem.","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\u002Ftemplates\u002Fprocesskit\u002Fv0.22.0\u002Fcontext\u002Fskills\u002Fprocesskit\u002Findex-management\u002FSKILL.md",[18,23],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000010","后端与接口","SCENARIO","使用场景",{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",[27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46],"management","index management","index","up","mcp server","for","errors","events","get","project","query","files","MCP","list","read","work","text","sqlite","server","search","2026-11-10 10:12:30",false,"---\nname: index-management\ndescription: |\n  SQLite-backed index over all entity files in the project. The read-side foundation for every other MCP server. Use whenever an agent needs to look up entities by ID, kind, state, or text — instead of grepping the filesystem.\nmetadata:\n  processkit:\n    apiVersion: processkit.projectious.work\u002Fv1\n    id: SKILL-index-management\n    version: \"1.0.0\"\n    created: 2026-04-06T00:00:00Z\n    category: processkit\n    layer: 0\n    provides:\n      primitives: []\n      mcp_tools: [reindex, query_entities, get_entity, search_entities, query_events, list_errors, \n            stats]\n---\n\n# Index Management\n\n## Intro\n\n`index-management` is the **read-side** foundation. It walks the project's\n`context\u002F` directory, parses every entity file, and writes a SQLite\ndatabase that other tools query. Its peer at Layer 0 is `id-management`\n(the write side, which allocates new IDs). Both are foundational to\nevery entity-creating skill.\n\n> **MCP server.** This skill ships a self-contained MCP server at\n> `mcp\u002Fserver.py` (PEP 723 script — requires `uv` and Python ≥ 3.10 on\n> PATH). Agent harnesses reach its tools by reading a single MCP config\n> file at startup, so the contents of `mcp\u002Fmcp-config.json` must be merged\n> into the harness's MCP config and placed at the harness-specific path\n> before this skill is usable. If processkit was installed by an installer,\n> that wiring is the installer's responsibility; if processkit was\n> installed manually, the project owner must do it by hand.\n\n## Overview\n\n### What it provides\n\n- **`reindex()`** — rebuild the SQLite index from scratch\n- **`query_entities(kind?, state?, limit?)`** — list entities matching filters\n- **`get_entity(id)`** — fetch a single entity by ID\n- **`search_entities(text, limit?)`** — full-text search across IDs, titles, bodies, specs\n- **`query_events(event_type?, subject?, actor?, limit?)`** — query the event log\n- **`list_errors()`** — files that failed to parse during the last reindex\n- **`stats()`** — counts of entities\u002Fevents\u002Ferrors in the index\n\n### When to use\n\nThe MCP server runs continuously inside the dev container. Agents call it\nwhenever they need fast lookup. Other MCP servers (`workitem-management`,\n`decision-record`, etc.) call `reindex` after writing a new entity to\nkeep the index fresh.\n\n### Where the database lives\n\n`\u003Cproject-root>\u002Fcontext\u002F.cache\u002Fprocesskit\u002Findex.sqlite`. Gitignored.\nRebuildable from source files at any time via `reindex()`.\n\n## Gotchas\n\nAgent-specific failure modes — provider-neutral pause-and-self-check items:\n\n- **Grepping the filesystem instead of calling `query_entities`.** The\n  index is faster, context-cheaper, and reflects parsed semantics\n  (state, kind, links) that grep can't see. Reach for `query_entities`\n  first; fall back to grep only if the query you need isn't supported.\n- **Trusting the index immediately after a hand-edit.** When you edit\n  an entity file directly (bypassing the MCP write tools), the index\n  is stale until the next `reindex()` call. Either go through the\n  write tools, or call `reindex()` yourself before querying.\n- **Forgetting to call `reindex` after writing a new entity.** Every\n  entity-creating MCP server should call `index_management.reindex`\n  (or the lighter `upsert_entity`) after a write. Not doing so means\n  the next query won't see the just-created entity.\n- **Treating index queries as transactional.** The index is eventually\n  consistent. A query immediately after a write may or may not return\n  the new row depending on whether reindex completed. If you need\n  strict consistency, reindex explicitly first.\n- **Trusting `list_errors()` as proof of correctness.** A clean errors\n  table means parsing succeeded, not that the entities are\n  semantically valid. Schema validation is a separate concern from\n  parse success.\n- **Using `search_entities` for exact-ID lookup.** `search_entities`\n  is full-text — it ranks by relevance and may miss exact ID matches\n  if the ID appears in many bodies. Use `get_entity(id)` for exact\n  lookup.\n- **Re-indexing inside a hot loop.** A full reindex walks the whole\n  `context\u002F` tree and is expensive. Batch your writes and reindex\n  once at the end, not after every entity.\n\n## Full reference\n\n### Database schema\n\nThree tables — see `src\u002Flib\u002Fprocesskit\u002Findex.py` for the DDL.\n\n```sql\nentities(id, kind, api_version, path, created, updated, title, state, labels_json, spec_json, body)\nevents(id, timestamp, event_type, actor, subject, subject_kind, summary, details_json, correlation_id, path)\nerrors(path, message)\n```\n\n`spec_json` holds the full `spec` block as JSON for queries we did not\nanticipate. `entities` and `events` overlap for LogEntry rows: a LogEntry\nappears in both, with the `events` table denormalizing the event-specific\nfields for fast filtering.\n\n### Reindex strategy\n\n`reindex()` is **destructive and atomic** — it deletes all rows and\nre-inserts. For typical projects this is fast (sub-second). For very\nlarge projects, an incremental update mode lands later (Phase 4+).\n\n### Errors table\n\nFiles that fail to parse get a row in the `errors` table instead of\ncrashing the reindex. `list_errors()` returns all such rows so the agent\ncan fix them. The errors table is cleared at the start of each reindex.\n\n### Tools that other servers call\n\n`workitem-management`, `decision-record`, `binding-management`, and\n`event-log` all call `index_management.upsert_entity` after writing a\nnew file. This keeps the index in sync without a full reindex on every\nmutation. From an MCP-protocol perspective each server has its own\nprocess — they communicate by sharing the same SQLite database file\n(WAL mode would be enabled in a future release for concurrent writes).\n\n### Configuration\n\nThe index database path is configurable via:\n\n- `processkit.toml` `[index] path = \"...\"` (relative to project root)\n- The `PROCESSKIT_INDEX_DB` environment variable\n- Default: `context\u002F.cache\u002Fprocesskit\u002Findex.sqlite` (gitignored cache)\n\n### Limitations at v0.3.0\n\n- **WAL mode enabled (v0.4.0+).** Multiple readers and a single writer\n  are safe; concurrent writes still serialize via WAL. Typical\n  AI-assisted sessions are single-writer anyway.\n- **No FTS.** Search uses `LIKE %text%`. SQLite FTS5 integration lands\n  in a later release.\n- **No incremental indexing.** Every reindex is a full sweep.\n","# Index Management\n\n## Intro\n\n`index-management` is the **read-side** foundation. It walks the project's\n`context\u002F` directory, parses every entity file, and writes a SQLite\ndatabase that other tools query. Its peer at Layer 0 is `id-management`\n(the write side, which allocates new IDs). Both are foundational to\nevery entity-creating skill.\n\n> **MCP server.** This skill ships a self-contained MCP server at\n> `mcp\u002Fserver.py` (PEP 723 script — requires `uv` and Python ≥ 3.10 on\n> PATH). Agent harnesses reach its tools by reading a single MCP config\n> file at startup, so the contents of `mcp\u002Fmcp-config.json` must be merged\n> into the harness's MCP config and placed at the harness-specific path\n> before this skill is usable. If processkit was installed by an installer,\n> that wiring is the installer's responsibility; if processkit was\n> installed manually, the project owner must do it by hand.\n\n## Overview\n\n### What it provides\n\n- **`reindex()`** — rebuild the SQLite index from scratch\n- **`query_entities(kind?, state?, limit?)`** — list entities matching filters\n- **`get_entity(id)`** — fetch a single entity by ID\n- **`search_entities(text, limit?)`** — full-text search across IDs, titles, bodies, specs\n- **`query_events(event_type?, subject?, actor?, limit?)`** — query the event log\n- **`list_errors()`** — files that failed to parse during the last reindex\n- **`stats()`** — counts of entities\u002Fevents\u002Ferrors in the index\n\n### When to use\n\nThe MCP server runs continuously inside the dev container. Agents call it\nwhenever they need fast lookup. Other MCP servers (`workitem-management`,\n`decision-record`, etc.) call `reindex` after writing a new entity to\nkeep the index fresh.\n\n### Where the database lives\n\n`\u003Cproject-root>\u002Fcontext\u002F.cache\u002Fprocesskit\u002Findex.sqlite`. Gitignored.\nRebuildable from source files at any time via `reindex()`.\n\n## Gotchas\n\nAgent-specific failure modes — provider-neutral pause-and-self-check items:\n\n- **Grepping the filesystem instead of calling `query_entities`.** The\n  index is faster, context-cheaper, and reflects parsed semantics\n  (state, kind, links) that grep can't see. Reach for `query_entities`\n  first; fall back to grep only if the query you need isn't supported.\n- **Trusting the index immediately after a hand-edit.** When you edit\n  an entity file directly (bypassing the MCP write tools), the index\n  is stale until the next `reindex()` call. Either go through the\n  write tools, or call `reindex()` yourself before querying.\n- **Forgetting to call `reindex` after writing a new entity.** Every\n  entity-creating MCP server should call `index_management.reindex`\n  (or the lighter `upsert_entity`) after a write. Not doing so means\n  the next query won't see the just-created entity.\n- **Treating index queries as transactional.** The index is eventually\n  consistent. A query immediately after a write may or may not return\n  the new row depending on whether reindex completed. If you need\n  strict consistency, reindex explicitly first.\n- **Trusting `list_errors()` as proof of correctness.** A clean errors\n  table means parsing succeeded, not that the entities are\n  semantically valid. Schema validation is a separate concern from\n  parse success.\n- **Using `search_entities` for exact-ID lookup.** `search_entities`\n  is full-text — it ranks by relevance and may miss exact ID matches\n  if the ID appears in many bodies. Use `get_entity(id)` for exact\n  lookup.\n- **Re-indexing inside a hot loop.** A full reindex walks the whole\n  `context\u002F` tree and is expensive. Batch your writes and reindex\n  once at the end, not after every entity.\n\n## Full reference\n\n### Database schema\n\nThree tables — see `src\u002Flib\u002Fprocesskit\u002Findex.py` for the DDL.\n\n```sql\nentities(id, kind, api_version, path, created, updated, title, state, labels_json, spec_json, body)\nevents(id, timestamp, event_type, actor, subject, subject_kind, summary, details_json, correlation_id, path)\nerrors(path, message)\n```\n\n`spec_json` holds the full `spec` block as JSON for queries we did not\nanticipate. `entities` and `events` overlap for LogEntry rows: a LogEntry\nappears in both, with the `events` table denormalizing the event-specific\nfields for fast filtering.\n\n### Reindex strategy\n\n`reindex()` is **destructive and atomic** — it deletes all rows and\nre-inserts. For typical projects this is fast (sub-second). For very\nlarge projects, an incremental update mode lands later (Phase 4+).\n\n### Errors table\n\nFiles that fail to parse get a row in the `errors` table instead of\ncrashing the reindex. `list_errors()` returns all such rows so the agent\ncan fix them. The errors table is cleared at the start of each reindex.\n\n### Tools that other servers call\n\n`workitem-management`, `decision-record`, `binding-management`, and\n`event-log` all call `index_management.upsert_entity` after writing a\nnew file. This keeps the index in sync without a full reindex on every\nmutation. From an MCP-protocol perspective each server has its own\nprocess — they communicate by sharing the same SQLite database file\n(WAL mode would be enabled in a future release for concurrent writes).\n\n### Configuration\n\nThe index database path is configurable via:\n\n- `processkit.toml` `[index] path = \"...\"` (relative to project root)\n- The `PROCESSKIT_INDEX_DB` environment variable\n- Default: `context\u002F.cache\u002Fprocesskit\u002Findex.sqlite` (gitignored cache)\n\n### Limitations at v0.3.0\n\n- **WAL mode enabled (v0.4.0+).** Multiple readers and a single writer\n  are safe; concurrent writes still serialize via WAL. Typical\n  AI-assisted sessions are single-writer anyway.\n- **No FTS.** Search uses `LIKE %text%`. SQLite FTS5 integration lands\n  in a later release.\n- **No incremental indexing.** Every reindex is a full sweep.\n",{"name":5,"metadata":52,"description":69},{"processkit":53},{"id":54,"layer":14,"created":55,"version":56,"category":57,"provides":58,"apiVersion":68},"SKILL-index-management",1775433600000,"1.0.0","processkit",{"mcp_tools":59,"primitives":67},[60,61,62,63,64,65,66],"reindex","query_entities","get_entity","search_entities","query_events","list_errors","stats",[],"processkit.projectious.work\u002Fv1","SQLite-backed index over all entity files in the project. The read-side foundation for every other MCP server. Use whenever an agent needs to look up entities by ID, kind, state, or text — instead of grepping the filesystem.\n","main","4876a92620ee792aac2ac9a1a806322d8a572367","archive\u002Fprocesskit-install-20261101\u002Fcontext\u002Ftemplates\u002Fprocesskit\u002Fv0.22.0\u002Fcontext\u002Fskills\u002Fprocesskit\u002Findex-management","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":78,"prompt":88,"zipAvailable":89},[79,82,85],{"tool":80,"command":81},"npx","npx skills add projectious-work\u002Fprocesskit --skill index-management",{"tool":83,"command":84},"bunx","bunx skills add projectious-work\u002Fprocesskit --skill index-management",{"tool":86,"command":87},"pnpm","pnpm dlx skills add projectious-work\u002Fprocesskit --skill index-management","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Farchive\u002Fprocesskit-install-20261101\u002Fcontext\u002Ftemplates\u002Fprocesskit\u002Fv0.22.0\u002Fcontext\u002Fskills\u002Fprocesskit\u002Findex-management\u002FSKILL.md",true,[91,97,103,107],{"path":92,"resourceType":93,"resourceTypeDesc":94,"sizeBytes":95,"sourceHtmlUrl":96},"config\u002Fsettings.toml","OTHER","其他",705,"https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Farchive\u002Fprocesskit-install-20261101\u002Fcontext\u002Ftemplates\u002Fprocesskit\u002Fv0.22.0\u002Fcontext\u002Fskills\u002Fprocesskit\u002Findex-management\u002Fconfig\u002Fsettings.toml",{"path":98,"resourceType":99,"resourceTypeDesc":100,"sizeBytes":101,"sourceHtmlUrl":102},"mcp\u002Fmcp-config.json","REFERENCE","参考文档",215,"https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Farchive\u002Fprocesskit-install-20261101\u002Fcontext\u002Ftemplates\u002Fprocesskit\u002Fv0.22.0\u002Fcontext\u002Fskills\u002Fprocesskit\u002Findex-management\u002Fmcp\u002Fmcp-config.json",{"path":104,"resourceType":99,"resourceTypeDesc":100,"sizeBytes":105,"sourceHtmlUrl":106},"mcp\u002FSERVER.md",2336,"https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Farchive\u002Fprocesskit-install-20261101\u002Fcontext\u002Ftemplates\u002Fprocesskit\u002Fv0.22.0\u002Fcontext\u002Fskills\u002Fprocesskit\u002Findex-management\u002Fmcp\u002FSERVER.md",{"path":108,"resourceType":109,"resourceTypeDesc":110,"sizeBytes":111,"sourceHtmlUrl":112},"mcp\u002Fserver.py","SCRIPT","脚本",6622,"https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Farchive\u002Fprocesskit-install-20261101\u002Fcontext\u002Ftemplates\u002Fprocesskit\u002Fv0.22.0\u002Fcontext\u002Fskills\u002Fprocesskit\u002Findex-management\u002Fmcp\u002Fserver.py",[114,121,128,134,154,173],{"skillNo":115,"name":116,"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":117,"categories":118,"keywords":119,"publishedAt":120},"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":122,"name":123,"nameZh":6,"description":124,"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":120},"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":129,"name":130,"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":131,"categories":132,"keywords":133,"publishedAt":120},"01M76650NN4T8P241021NBR64S","standard-agent-building","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fstandard-agent-building\u002FSKILL.md",[],[],{"skillNo":135,"name":136,"nameZh":6,"description":137,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":138,"categories":139,"keywords":141,"publishedAt":47},"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",[140],{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},[27,142,143,144,145,146,42,147,35,148,149,150,136,40,151,152,153],"skill management","ontology","v1","create","as","transition","manage","update","alpha","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":47},"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":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},[162,163,27,164,165,166,167,168,169,39,170,171,42,35,148,150,172],"schema","schema management","contract","rebuild","contracts","workflows","wed","generated","validation","deterministic","mode",{"skillNo":174,"name":175,"nameZh":6,"description":176,"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":179,"publishedAt":47},"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",[],[180,27,181,182,145,183,184,185,42,35,186,187,149,188,150,29,189,37,190,191],"proposition management","proposition","interface","claims","record","world","creation","event","WSJF","updates","facts","log"]