[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6NPZKVR6J37SJDPFPVPG8T4":3,"skill-same:01M6NPZKVR6J37SJDPFPVPG8T4":81},{"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":43,"linkOnly":44,"markdownContent":45,"markdownBody":46,"frontMatter":47,"refName":55,"commitSha":56,"skillDirectory":57,"forksCount":14,"repositoryHtmlUrl":58,"repositoryHomepageUrl":59,"ownerHtmlUrl":60,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":61,"install":62,"resources":75},"01M6NPZKVR6J37SJDPFPVPG8T4","graphql-patterns",null,"GraphQL schema design, resolver patterns, DataLoader, pagination, federation, and schema evolution. Use when designing a GraphQL schema, implementing resolvers, diagnosing N+1 problems, adding pagination, setting up federation, or evolving a schema without breaking clients.","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\u002Fengineering\u002Fgraphql-patterns\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000015","视觉设计","SCENARIO","使用场景",[24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42],"graphql patterns","GraphQL","patterns","schema design","implementing","resolvers","resolver","graphql schema design","graphql schema","clients","pagination","design","up","schema","use","evolution","work","federation","engineering","2026-11-10 10:12:30",false,"---\nname: graphql-patterns\ndescription: |\n  GraphQL schema design, resolver patterns, DataLoader, pagination, federation, and schema evolution. Use when designing a GraphQL schema, implementing resolvers, diagnosing N+1 problems, adding pagination, setting up federation, or evolving a schema without breaking clients.\nmetadata:\n  processkit:\n    apiVersion: processkit.projectious.work\u002Fv2\n    id: SKILL-graphql-patterns\n    version: \"1.1.0\"\n    created: 2026-04-06T00:00:00Z\n    category: engineering\n---\n\n# GraphQL Patterns\n\n## Intro\n\nGraphQL schemas are designed from the client's perspective, not the\ndatabase. Resolvers stay thin, DataLoader prevents N+1 fan-out, lists\nuse Relay-style cursor pagination, and schemas evolve only by adding\nfields and deprecating the old ones.\n\n## Overview\n\n### Schema design\n\nModel the domain as the client wants to consume it. Types are nouns\n(`User`, `Order`); queries are reads (`user(id: ID!)`,\n`orders(filter: OrderFilter)`); mutations are verb-object writes\n(`createOrder`, `cancelOrder`); subscriptions are real-time streams\n(`orderStatusChanged(orderId: ID!)`).\n\nDefault fields to non-nullable (`String!`) and only mark them\nnullable when null is a meaningful state — the inverse of most REST\nAPIs. Define custom scalars for domain values like `DateTime` and\n`Email`. Always use `input` types for mutation arguments so they can\nevolve independently of output types.\n\n### Resolver patterns\n\n- **Root resolvers** handle top-level Query and Mutation fields.\n- **Field resolvers** handle nested fields that come from a different\n  source than the parent.\n- **Default resolvers** return `parent[fieldName]` automatically — do\n  not write trivial passthrough resolvers.\n\nResolvers stay thin: they call into a service or domain layer. Never\nembed business logic, validation, or SQL inside a resolver.\n\n### DataLoader and N+1\n\nA list query that triggers one extra query per item is the N+1\nproblem. Solve it with DataLoader, which batches and caches reads\nwithin a single request:\n\n```javascript\nconst orderLoader = new DataLoader(async (userIds) => {\n  const orders = await db.orders.findByUserIds(userIds);\n  return userIds.map(id => orders.filter(o => o.userId === id));\n});\n\u002F\u002F In User.orders resolver: (user) => orderLoader.load(user.id)\n```\n\nTwo non-negotiable rules:\n\n1. Create a new DataLoader instance **per request**. Sharing across\n   requests leaks data between users.\n2. The batch function must return results in the **same order** as\n   the input keys, with `null` for missing entries.\n\n### Pagination (Relay Connection spec)\n\nUse cursor-based connections for every list field:\n\n```graphql\ntype OrderConnection {\n  edges: [OrderEdge!]!\n  pageInfo: PageInfo!\n}\ntype OrderEdge   { cursor: String!  node: Order! }\ntype PageInfo    { hasNextPage: Boolean!  hasPreviousPage: Boolean!  endCursor: String }\n```\n\nCursors are opaque base64 tokens. Default `first` to 20, enforce a\nhard maximum of 100 to prevent unbounded queries.\n\n### Errors\n\nPrefer **typed union results** for expected, domain-level errors so\nclients pattern-match in a type-safe way:\n\n```graphql\nunion CreateOrderResult = Order | ValidationError | InsufficientStock\n```\n\nReserve the top-level `errors` array for unexpected failures\n(authentication, server errors, transport). Always include\n`extensions.code` (e.g. `UNAUTHENTICATED`, `INTERNAL_SERVER_ERROR`)\nso clients can branch on a stable string.\n\n### Schema evolution\n\n- **Adding** types, fields, and enum values is always safe.\n- **Deprecating**: mark with `@deprecated(reason: \"Use newField\")`,\n  monitor field-usage telemetry, then remove once usage is near zero.\n- **Never** change a field's type, and never make a nullable field\n  non-nullable — both are silently breaking.\n\n## Gotchas\n\nAgent-specific failure modes — provider-neutral pause-and-self-check items:\n\n- **DataLoader shared across requests.** A DataLoader caches results for the lifetime of the instance. Sharing one instance across requests means user A's query result can be returned to user B. Always create a new DataLoader instance per request.\n- **N+1 resolvers without DataLoader.** A list resolver that triggers one database query per item will fan out to N+1 queries for a list of N items. This is the most common GraphQL performance failure mode. Every nested field resolver that fetches from a database or external service needs to be batched with DataLoader.\n- **Making nullable fields non-nullable in a schema update.** A nullable field made non-nullable is a breaking change for any client that previously received `null` and handled it. Schema evolution only goes in one direction: add fields, deprecate old ones, never tighten nullability.\n- **Returning `Boolean` from mutations.** A mutation that returns `true\u002Ffalse` forces the client to re-fetch the entity to update its cache. Return the affected entity or a result union so the client has the data it needs without a round-trip.\n- **No depth or complexity limits in production.** A deeply nested recursive query can cause exponential database fan-out. Without depth and complexity limits, a single malicious or buggy query can exhaust the server. Enable depth limiting and cost-based rate limiting.\n- **CRUD resolvers that mirror the database schema.** GraphQL APIs should be designed from the client's consumption needs, not from the storage model. A mutation called `updateOrder(id, status, tracking_number, shipped_at, ...)` with 10 optional fields is a database operation disguised as an intent; model it as `shipOrder(id, tracking_number)` instead.\n- **One mutation that does everything with optional fields.** Large mutations with many optional fields make it impossible for the client to know which fields are required for which operations and make server-side validation complex. Split into intent-specific, focused mutations.\n\n## Full reference\n\n### Federation\n\nFor microservice architectures using Apollo Federation:\n\n- Each service owns a subgraph and marks shared entities with\n  `@key(fields: \"id\")`.\n- The gateway composes subgraphs into a supergraph at build time, not\n  at request time, so composition errors fail fast in CI.\n- A subgraph only extends fields it can resolve from its own data —\n  never reach across subgraphs from a resolver.\n\nThis keeps each subgraph independently deployable and avoids the\ndistributed-monolith trap.\n\n### Persisted queries\n\nIn production, prefer **automatic persisted queries (APQ)**: the\nclient first sends only the SHA-256 hash of the query; the server\nlooks it up and executes the cached query. Benefits:\n\n- Reduced bandwidth (hash, not full query string).\n- Allowlisting becomes possible — reject any query whose hash is not\n  in the registry, blocking arbitrary or hostile queries.\n- Easier query-level caching at the edge.\n\n### Performance and depth limits\n\nReject queries that exceed a sensible depth (e.g. 10) and complexity\nscore. Without limits, a malicious client can request a deeply\nrecursive graph and exhaust the server. Most GraphQL servers ship a\ndepth-limit middleware — turn it on.\n\nRate-limit by query cost, not by request count. A single GraphQL\nrequest can be cheap or catastrophically expensive depending on the\nselection set.\n\n### Subscriptions\n\nSubscriptions push events over WebSockets. Use them sparingly: each\nsubscription holds a connection, so the server pays for every\nconnected client. Prefer polling for low-frequency updates and\nreserve subscriptions for genuinely real-time UX (live order status,\nchat, collaborative editing).\n\n### Anti-patterns to avoid\n\n- **CRUD resolvers that mirror database tables.** Design from the\n  client's needs, not from the schema migration file.\n- **One mutation that does everything.** Split into named\n  intent-specific mutations (`shipOrder`, `cancelOrder`) instead of\n  `updateOrder(input)` with twenty optional fields.\n- **Returning `Boolean` from mutations.** Return the affected entity\n  or a result union so the client can update its cache.\n- **DataLoader shared across requests.** Per-request, always.\n- **Throwing strings or generic `Error` from resolvers.** Use the\n  result-union pattern for expected errors.\n- **Letting clients send arbitrary queries in production.** Use APQ\n  + allowlisting.\n\n### When to break the rules\n\nInternal-only APIs with a single trusted client can skip persisted\nqueries, depth limits, and federation. Tiny schemas (under ~20 types)\ncan skip DataLoader if every resolver hits an in-memory cache. The\nresult-union error pattern adds verbosity that may not be worth it\nfor a small admin UI — top-level `errors` is fine there. As always,\nunderstand the rule first; only then break it deliberately.\n","# GraphQL Patterns\n\n## Intro\n\nGraphQL schemas are designed from the client's perspective, not the\ndatabase. Resolvers stay thin, DataLoader prevents N+1 fan-out, lists\nuse Relay-style cursor pagination, and schemas evolve only by adding\nfields and deprecating the old ones.\n\n## Overview\n\n### Schema design\n\nModel the domain as the client wants to consume it. Types are nouns\n(`User`, `Order`); queries are reads (`user(id: ID!)`,\n`orders(filter: OrderFilter)`); mutations are verb-object writes\n(`createOrder`, `cancelOrder`); subscriptions are real-time streams\n(`orderStatusChanged(orderId: ID!)`).\n\nDefault fields to non-nullable (`String!`) and only mark them\nnullable when null is a meaningful state — the inverse of most REST\nAPIs. Define custom scalars for domain values like `DateTime` and\n`Email`. Always use `input` types for mutation arguments so they can\nevolve independently of output types.\n\n### Resolver patterns\n\n- **Root resolvers** handle top-level Query and Mutation fields.\n- **Field resolvers** handle nested fields that come from a different\n  source than the parent.\n- **Default resolvers** return `parent[fieldName]` automatically — do\n  not write trivial passthrough resolvers.\n\nResolvers stay thin: they call into a service or domain layer. Never\nembed business logic, validation, or SQL inside a resolver.\n\n### DataLoader and N+1\n\nA list query that triggers one extra query per item is the N+1\nproblem. Solve it with DataLoader, which batches and caches reads\nwithin a single request:\n\n```javascript\nconst orderLoader = new DataLoader(async (userIds) => {\n  const orders = await db.orders.findByUserIds(userIds);\n  return userIds.map(id => orders.filter(o => o.userId === id));\n});\n\u002F\u002F In User.orders resolver: (user) => orderLoader.load(user.id)\n```\n\nTwo non-negotiable rules:\n\n1. Create a new DataLoader instance **per request**. Sharing across\n   requests leaks data between users.\n2. The batch function must return results in the **same order** as\n   the input keys, with `null` for missing entries.\n\n### Pagination (Relay Connection spec)\n\nUse cursor-based connections for every list field:\n\n```graphql\ntype OrderConnection {\n  edges: [OrderEdge!]!\n  pageInfo: PageInfo!\n}\ntype OrderEdge   { cursor: String!  node: Order! }\ntype PageInfo    { hasNextPage: Boolean!  hasPreviousPage: Boolean!  endCursor: String }\n```\n\nCursors are opaque base64 tokens. Default `first` to 20, enforce a\nhard maximum of 100 to prevent unbounded queries.\n\n### Errors\n\nPrefer **typed union results** for expected, domain-level errors so\nclients pattern-match in a type-safe way:\n\n```graphql\nunion CreateOrderResult = Order | ValidationError | InsufficientStock\n```\n\nReserve the top-level `errors` array for unexpected failures\n(authentication, server errors, transport). Always include\n`extensions.code` (e.g. `UNAUTHENTICATED`, `INTERNAL_SERVER_ERROR`)\nso clients can branch on a stable string.\n\n### Schema evolution\n\n- **Adding** types, fields, and enum values is always safe.\n- **Deprecating**: mark with `@deprecated(reason: \"Use newField\")`,\n  monitor field-usage telemetry, then remove once usage is near zero.\n- **Never** change a field's type, and never make a nullable field\n  non-nullable — both are silently breaking.\n\n## Gotchas\n\nAgent-specific failure modes — provider-neutral pause-and-self-check items:\n\n- **DataLoader shared across requests.** A DataLoader caches results for the lifetime of the instance. Sharing one instance across requests means user A's query result can be returned to user B. Always create a new DataLoader instance per request.\n- **N+1 resolvers without DataLoader.** A list resolver that triggers one database query per item will fan out to N+1 queries for a list of N items. This is the most common GraphQL performance failure mode. Every nested field resolver that fetches from a database or external service needs to be batched with DataLoader.\n- **Making nullable fields non-nullable in a schema update.** A nullable field made non-nullable is a breaking change for any client that previously received `null` and handled it. Schema evolution only goes in one direction: add fields, deprecate old ones, never tighten nullability.\n- **Returning `Boolean` from mutations.** A mutation that returns `true\u002Ffalse` forces the client to re-fetch the entity to update its cache. Return the affected entity or a result union so the client has the data it needs without a round-trip.\n- **No depth or complexity limits in production.** A deeply nested recursive query can cause exponential database fan-out. Without depth and complexity limits, a single malicious or buggy query can exhaust the server. Enable depth limiting and cost-based rate limiting.\n- **CRUD resolvers that mirror the database schema.** GraphQL APIs should be designed from the client's consumption needs, not from the storage model. A mutation called `updateOrder(id, status, tracking_number, shipped_at, ...)` with 10 optional fields is a database operation disguised as an intent; model it as `shipOrder(id, tracking_number)` instead.\n- **One mutation that does everything with optional fields.** Large mutations with many optional fields make it impossible for the client to know which fields are required for which operations and make server-side validation complex. Split into intent-specific, focused mutations.\n\n## Full reference\n\n### Federation\n\nFor microservice architectures using Apollo Federation:\n\n- Each service owns a subgraph and marks shared entities with\n  `@key(fields: \"id\")`.\n- The gateway composes subgraphs into a supergraph at build time, not\n  at request time, so composition errors fail fast in CI.\n- A subgraph only extends fields it can resolve from its own data —\n  never reach across subgraphs from a resolver.\n\nThis keeps each subgraph independently deployable and avoids the\ndistributed-monolith trap.\n\n### Persisted queries\n\nIn production, prefer **automatic persisted queries (APQ)**: the\nclient first sends only the SHA-256 hash of the query; the server\nlooks it up and executes the cached query. Benefits:\n\n- Reduced bandwidth (hash, not full query string).\n- Allowlisting becomes possible — reject any query whose hash is not\n  in the registry, blocking arbitrary or hostile queries.\n- Easier query-level caching at the edge.\n\n### Performance and depth limits\n\nReject queries that exceed a sensible depth (e.g. 10) and complexity\nscore. Without limits, a malicious client can request a deeply\nrecursive graph and exhaust the server. Most GraphQL servers ship a\ndepth-limit middleware — turn it on.\n\nRate-limit by query cost, not by request count. A single GraphQL\nrequest can be cheap or catastrophically expensive depending on the\nselection set.\n\n### Subscriptions\n\nSubscriptions push events over WebSockets. Use them sparingly: each\nsubscription holds a connection, so the server pays for every\nconnected client. Prefer polling for low-frequency updates and\nreserve subscriptions for genuinely real-time UX (live order status,\nchat, collaborative editing).\n\n### Anti-patterns to avoid\n\n- **CRUD resolvers that mirror database tables.** Design from the\n  client's needs, not from the schema migration file.\n- **One mutation that does everything.** Split into named\n  intent-specific mutations (`shipOrder`, `cancelOrder`) instead of\n  `updateOrder(input)` with twenty optional fields.\n- **Returning `Boolean` from mutations.** Return the affected entity\n  or a result union so the client can update its cache.\n- **DataLoader shared across requests.** Per-request, always.\n- **Throwing strings or generic `Error` from resolvers.** Use the\n  result-union pattern for expected errors.\n- **Letting clients send arbitrary queries in production.** Use APQ\n  + allowlisting.\n\n### When to break the rules\n\nInternal-only APIs with a single trusted client can skip persisted\nqueries, depth limits, and federation. Tiny schemas (under ~20 types)\ncan skip DataLoader if every resolver hits an in-memory cache. The\nresult-union error pattern adds verbosity that may not be worth it\nfor a small admin UI — top-level `errors` is fine there. As always,\nunderstand the rule first; only then break it deliberately.\n",{"name":5,"metadata":48,"description":54},{"processkit":49},{"id":50,"created":51,"version":52,"category":42,"apiVersion":53},"SKILL-graphql-patterns",1775433600000,"1.1.0","processkit.projectious.work\u002Fv2","GraphQL schema design, resolver patterns, DataLoader, pagination, federation, and schema evolution. Use when designing a GraphQL schema, implementing resolvers, diagnosing N+1 problems, adding pagination, setting up federation, or evolving a schema without breaking clients.\n","main","7231f100e266e30d2c1699ecde5c00602471dc3c","src\u002Fcontext\u002Fskills\u002Fengineering\u002Fgraphql-patterns","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":63,"prompt":73,"zipAvailable":74},[64,67,70],{"tool":65,"command":66},"npx","npx skills add projectious-work\u002Fprocesskit --skill graphql-patterns",{"tool":68,"command":69},"bunx","bunx skills add projectious-work\u002Fprocesskit --skill graphql-patterns",{"tool":71,"command":72},"pnpm","pnpm dlx skills add projectious-work\u002Fprocesskit --skill graphql-patterns","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Fsrc\u002Fcontext\u002Fskills\u002Fengineering\u002Fgraphql-patterns\u002FSKILL.md",true,[76],{"path":77,"resourceType":78,"resourceTypeDesc":79,"sizeBytes":14,"sourceHtmlUrl":80},"scripts\u002F.gitkeep","SCRIPT","脚本","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Fsrc\u002Fcontext\u002Fskills\u002Fengineering\u002Fgraphql-patterns\u002Fscripts\u002F.gitkeep",[82,89,96,102,127,146],{"skillNo":83,"name":84,"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":85,"categories":86,"keywords":87,"publishedAt":88},"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":90,"name":91,"nameZh":6,"description":92,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":93,"categories":94,"keywords":95,"publishedAt":88},"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":97,"name":98,"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":99,"categories":100,"keywords":101,"publishedAt":88},"01M76650NN4T8P241021NBR64S","standard-agent-building","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fstandard-agent-building\u002FSKILL.md",[],[],{"skillNo":103,"name":104,"nameZh":6,"description":105,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":106,"categories":107,"keywords":111,"publishedAt":43},"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",[108],{"categoryNo":109,"categoryName":110,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",[112,113,114,115,116,117,40,118,119,120,121,122,104,123,124,125,126],"management","skill management","ontology","v1","create","as","transition","get","manage","update","alpha","list","package","skill package","manifests",{"skillNo":128,"name":129,"nameZh":6,"description":130,"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":134,"publishedAt":43},"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",[133],{"categoryNo":109,"categoryName":110,"taxonomyType":21,"taxonomyTypeDesc":22},[37,135,112,136,137,138,139,140,141,142,143,144,40,119,120,122,145],"schema management","contract","rebuild","contracts","workflows","wed","generated","MCP","validation","deterministic","mode",{"skillNo":147,"name":148,"nameZh":6,"description":149,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":150,"categories":151,"keywords":152,"publishedAt":43},"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",[],[153,112,154,155,116,156,157,158,40,119,159,160,121,161,122,162,163,164,165,166],"proposition management","proposition","interface","claims","record","world","creation","event","WSJF","index","updates","query","facts","log"]