[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M79Z6YAE80FMSP7SB81PG8XW":3,"skill-same:01M79Z6YAE80FMSP7SB81PG8XW":52},{"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":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":31,"commitSha":32,"skillDirectory":33,"forksCount":14,"repositoryHtmlUrl":34,"repositoryHomepageUrl":6,"ownerHtmlUrl":35,"ownerBlogUrl":36,"ownerDescription":37,"updatedAt":38,"install":39,"resources":51},"01M79Z6YAE80FMSP7SB81PG8XW","go-graphql",null,"Build GraphQL servers in Go with gqlgen: schema-first generation, resolver structure, the N+1 problem and dataloaders, complexity and depth limits, error presentation, field-level authorization, and testing resolvers. Use when implementing or reviewing a GraphQL API, when a query fans out into hundreds of database calls, or when deciding what a resolver may expose. Trigger examples: \"GraphQL\", \"gqlgen\", \"resolver\", \"N+1 queries\", \"dataloader\", \"query complexity limit\", \"GraphQL schema\". Not for: REST and OpenAPI (go-openapi), gRPC (go-grpc), general HTTP middleware and shutdown (go-api-design).","kongnakornna\u002Fgolangapi","kongnakornna","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F10528904?v=4","01M79Z5JQPPV5CWQHXS235Q9MD","Kongnakorn Jantakun","01M79Z5JRHJJJ6PR2DGQS4P985",0,"MIT","https:\u002F\u002Fgithub.com\u002Fkongnakornna\u002Fgolangapi\u002Fblob\u002Fmain\u002F.github\u002Fskills\u002Fgo-agent-skills\u002Fskills\u002F(architecture)\u002Fgo-graphql\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: go-graphql\ndescription: >\n  Build GraphQL servers in Go with gqlgen: schema-first generation, resolver\n  structure, the N+1 problem and dataloaders, complexity and depth limits,\n  error presentation, field-level authorization, and testing resolvers. Use\n  when implementing or reviewing a GraphQL API, when a query fans out into\n  hundreds of database calls, or when deciding what a resolver may expose.\n  Trigger examples: \"GraphQL\", \"gqlgen\", \"resolver\", \"N+1 queries\",\n  \"dataloader\", \"query complexity limit\", \"GraphQL schema\".\n  Not for: REST and OpenAPI (go-openapi), gRPC (go-grpc), general HTTP\n  middleware and shutdown (go-api-design).\nuser-invocable: true\nlicense: MIT\ncompatibility: Designed for Claude Code or similar AI coding agents working on Go projects. Requires the Go toolchain. gqlgen is installed as a module tool.\nallowed-tools: Read Edit Write Glob Grep Bash(go:*) Bash(gofmt:*) Bash(gqlgen:*)\nmetadata:\n  author: eduardo-sl\n  version: \"1.0.1\"\n---\n\n# Go GraphQL\n\nGraphQL moves query planning to the client. That is the feature and the\ndanger: one innocuous query can become ten thousand database round trips, and\none over-permissive field can leak another tenant's data. Both are solved at\nthe server, not in the schema review.\n\n## 1. Schema-First with gqlgen\n\nThe `.graphql` schema is the source of truth. gqlgen generates models,\nresolver stubs, and the execution layer from it.\n\n```bash\ngo get -tool github.com\u002F99designs\u002Fgqlgen\ngo tool gqlgen init      # once\ngo tool gqlgen generate  # after every schema change\n```\n\n```yaml\n# gqlgen.yml — bind generated types to your own models\nmodels:\n  User:\n    model: github.com\u002Fmyorg\u002Fapp\u002Finternal\u002Fdomain.User\n  ID:\n    model:\n      - github.com\u002F99designs\u002Fgqlgen\u002Fgraphql.ID\n      - github.com\u002F99designs\u002Fgqlgen\u002Fgraphql.Int64\n```\n\nBind domain types explicitly. Left to itself gqlgen generates a parallel set\nof anaemic structs, and every resolver becomes a mapping function.\n\nCommit generated code, and fail CI when it is stale:\n\n```bash\ngo tool gqlgen generate && git diff --exit-code\n```\n\nNever edit `generated.go` or `models_gen.go`. `resolver.go` and the\n`*.resolvers.go` files are yours.\n\n## 2. Resolvers Stay Thin\n\nA resolver translates a GraphQL request into a service call. It contains no\nbusiness logic and no SQL.\n\n```go\nfunc (r *queryResolver) User(ctx context.Context, id string) (*domain.User, error) {\n    u, err := r.users.Find(ctx, id)\n    if errors.Is(err, domain.ErrNotFound) {\n        return nil, nil \u002F\u002F nullable field: absent, not an error\n    }\n    if err != nil {\n        return nil, fmt.Errorf(\"find user %s: %w\", id, err)\n    }\n    return u, nil\n}\n```\n\nInject dependencies through the `Resolver` struct, never through package\nglobals:\n\n```go\ntype Resolver struct {\n    users  UserService\n    orders OrderService\n    loader *Loaders\n}\n```\n\nAlways propagate `ctx`. It carries the request deadline, the authenticated\nprincipal, and the per-request dataloaders.\n\n## 3. The N+1 Problem — the one that matters\n\nA field resolver on a list type runs once per element.\n\n```go\n\u002F\u002F ❌ 1 query for the orders, then N queries for the users\nfunc (r *orderResolver) Customer(ctx context.Context, obj *domain.Order) (*domain.User, error) {\n    return r.users.Find(ctx, obj.CustomerID)\n}\n```\n\nBatch with a dataloader. It collects the keys requested within a short window\nand issues one query.\n\n```go\nimport \"github.com\u002Fvikstrous\u002Fdataloadgen\"\n\ntype Loaders struct {\n    UserByID *dataloadgen.Loader[string, *domain.User]\n}\n\nfunc NewLoaders(s UserService) *Loaders {\n    return &Loaders{\n        UserByID: dataloadgen.NewLoader(func(ctx context.Context, ids []string) ([]*domain.User, []error) {\n            return s.FindMany(ctx, ids) \u002F\u002F ONE query for all ids\n        }, dataloadgen.WithWait(time.Millisecond)),\n    }\n}\n\n\u002F\u002F ✅ 1 query for the orders, 1 for all customers\nfunc (r *orderResolver) Customer(ctx context.Context, obj *domain.Order) (*domain.User, error) {\n    return loadersFrom(ctx).UserByID.Load(ctx, obj.CustomerID)\n}\n```\n\nLoaders are **per request**, installed by middleware. A process-wide loader\ncaches across users and leaks data between them.\n\n```go\nfunc withLoaders(svc UserService, next http.Handler) http.Handler {\n    return http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {\n        ctx := context.WithValue(r.Context(), loadersKey{}, NewLoaders(svc))\n        next.ServeHTTP(w, r.WithContext(ctx))\n    })\n}\n```\n\nThe batch function must return results **in the order of the keys it was\ngiven**, with a nil entry and an error per missing key. Returning a shorter\nslice silently misaligns every result.\n\n## 4. Bound Every Query\n\nA public GraphQL endpoint without limits is a denial-of-service endpoint.\n\n```go\nsrv := handler.New(generated.NewExecutableSchema(cfg))\nsrv.AddTransport(transport.POST{})\nsrv.SetQueryCache(lru.New[*ast.QueryDocument](1000))\nsrv.Use(extension.FixedComplexityLimit(300))\nsrv.Use(extension.AutomaticPersistedQuery{Cache: lru.New[string](100)})\n```\n\n- **Complexity limit** — assign a cost per field, higher for list fields with\n  a large `first`. Start at a number your slowest legitimate query fits under,\n  then measure.\n- **Depth** — recursive types (`user { orders { customer { orders ... } } }`)\n  must be bounded. gqlgen has no built-in depth limit; enforce it in an\n  operation middleware.\n- **Pagination is mandatory** on every list field. A field returning an\n  unbounded list is a schema bug.\n- **Introspection** is only enabled if you install `extension.Introspection`.\n  Do not install it in production, or gate it behind an authenticated role.\n- **Persisted queries** let a public client send a hash instead of a document,\n  so the server executes only queries you shipped.\n\nSet `srv.AroundOperations` to enforce a per-operation timeout, and always run\nbehind an `http.Server` with `ReadTimeout` and `WriteTimeout` set.\n\n## 5. Errors\n\nGraphQL returns 200 with an `errors` array. Never leak internals into it.\n\n```go\nsrv.SetErrorPresenter(func(ctx context.Context, e error) *gqlerror.Error {\n    err := graphql.DefaultErrorPresenter(ctx, e)\n\n    var domainErr *domain.ValidationError\n    if errors.As(e, &domainErr) {\n        err.Message = domainErr.Message\n        err.Extensions = map[string]any{\"code\": \"VALIDATION_FAILED\"}\n        return err\n    }\n\n    slog.ErrorContext(ctx, \"graphql resolver failed\", \"error\", e)\n    err.Message = \"internal server error\" \u002F\u002F stable, safe\n    err.Extensions = map[string]any{\"code\": \"INTERNAL\"}\n    return err\n})\n```\n\nUse `srv.SetRecoverFunc` to convert a resolver panic into an error instead of\nkilling the connection, and log it with the stack.\n\nRemember the nullability rule: an error on a non-null field nulls out its\nnearest nullable ancestor. Make a field non-null only when it can never\nlegitimately be absent.\n\n## 6. Authorization Belongs on the Field\n\nObject-level checks are not enough — a client can reach an object through\nseveral paths.\n\n```graphql\ndirective @hasRole(role: Role!) on FIELD_DEFINITION\n\ntype User {\n  id: ID!\n  email: String! @hasRole(role: ADMIN)\n}\n```\n\n```go\ncfg.Directives.HasRole = func(ctx context.Context, obj any, next graphql.Resolver, role model.Role) (any, error) {\n    if !auth.FromContext(ctx).HasRole(role) {\n        return nil, gqlerror.Errorf(\"access denied\")\n    }\n    return next(ctx)\n}\n```\n\nAuthenticate in HTTP middleware, before the GraphQL handler. Authorize in the\ndirective or the resolver, using the principal from the context — never from\na query argument.\n\n## 7. Testing\n\n```go\nfunc TestUserQuery(t *testing.T) {\n    c := client.New(handler.NewDefaultServer(generated.NewExecutableSchema(cfg)))\n\n    var resp struct {\n        User struct{ ID, Email string }\n    }\n    c.MustPost(`{ user(id: \"u-1\") { id email } }`, &resp)\n\n    require.Equal(t, \"u-1\", resp.User.ID)\n}\n```\n\nAssert the query count for any resolver with a dataloader — that is the only\nway an N+1 regression fails a build rather than a dashboard:\n\n```go\nrequire.Equal(t, 2, db.QueryCount(), \"expected batched loads, got N+1\")\n```\n\n## Verification Checklist\n\n1. Schema is the source of truth; generated files are committed and CI-checked\n2. Generated types bind to domain models via `gqlgen.yml`\n3. Resolvers contain no business logic and always propagate `ctx`\n4. Every list-field resolver that fetches by ID goes through a dataloader\n5. Dataloaders are constructed per request, never shared across requests\n6. Batch functions return one result per key, in key order\n7. A complexity limit and a depth bound are configured and tested\n8. Every list field is paginated\n9. Introspection is disabled or role-gated in production\n10. An error presenter strips internal errors; a recover func is installed\n11. Authorization is enforced per field, from the context principal\n12. A test asserts the query count for at least one batched field\n","# Go GraphQL\n\nGraphQL moves query planning to the client. That is the feature and the\ndanger: one innocuous query can become ten thousand database round trips, and\none over-permissive field can leak another tenant's data. Both are solved at\nthe server, not in the schema review.\n\n## 1. Schema-First with gqlgen\n\nThe `.graphql` schema is the source of truth. gqlgen generates models,\nresolver stubs, and the execution layer from it.\n\n```bash\ngo get -tool github.com\u002F99designs\u002Fgqlgen\ngo tool gqlgen init      # once\ngo tool gqlgen generate  # after every schema change\n```\n\n```yaml\n# gqlgen.yml — bind generated types to your own models\nmodels:\n  User:\n    model: github.com\u002Fmyorg\u002Fapp\u002Finternal\u002Fdomain.User\n  ID:\n    model:\n      - github.com\u002F99designs\u002Fgqlgen\u002Fgraphql.ID\n      - github.com\u002F99designs\u002Fgqlgen\u002Fgraphql.Int64\n```\n\nBind domain types explicitly. Left to itself gqlgen generates a parallel set\nof anaemic structs, and every resolver becomes a mapping function.\n\nCommit generated code, and fail CI when it is stale:\n\n```bash\ngo tool gqlgen generate && git diff --exit-code\n```\n\nNever edit `generated.go` or `models_gen.go`. `resolver.go` and the\n`*.resolvers.go` files are yours.\n\n## 2. Resolvers Stay Thin\n\nA resolver translates a GraphQL request into a service call. It contains no\nbusiness logic and no SQL.\n\n```go\nfunc (r *queryResolver) User(ctx context.Context, id string) (*domain.User, error) {\n    u, err := r.users.Find(ctx, id)\n    if errors.Is(err, domain.ErrNotFound) {\n        return nil, nil \u002F\u002F nullable field: absent, not an error\n    }\n    if err != nil {\n        return nil, fmt.Errorf(\"find user %s: %w\", id, err)\n    }\n    return u, nil\n}\n```\n\nInject dependencies through the `Resolver` struct, never through package\nglobals:\n\n```go\ntype Resolver struct {\n    users  UserService\n    orders OrderService\n    loader *Loaders\n}\n```\n\nAlways propagate `ctx`. It carries the request deadline, the authenticated\nprincipal, and the per-request dataloaders.\n\n## 3. The N+1 Problem — the one that matters\n\nA field resolver on a list type runs once per element.\n\n```go\n\u002F\u002F ❌ 1 query for the orders, then N queries for the users\nfunc (r *orderResolver) Customer(ctx context.Context, obj *domain.Order) (*domain.User, error) {\n    return r.users.Find(ctx, obj.CustomerID)\n}\n```\n\nBatch with a dataloader. It collects the keys requested within a short window\nand issues one query.\n\n```go\nimport \"github.com\u002Fvikstrous\u002Fdataloadgen\"\n\ntype Loaders struct {\n    UserByID *dataloadgen.Loader[string, *domain.User]\n}\n\nfunc NewLoaders(s UserService) *Loaders {\n    return &Loaders{\n        UserByID: dataloadgen.NewLoader(func(ctx context.Context, ids []string) ([]*domain.User, []error) {\n            return s.FindMany(ctx, ids) \u002F\u002F ONE query for all ids\n        }, dataloadgen.WithWait(time.Millisecond)),\n    }\n}\n\n\u002F\u002F ✅ 1 query for the orders, 1 for all customers\nfunc (r *orderResolver) Customer(ctx context.Context, obj *domain.Order) (*domain.User, error) {\n    return loadersFrom(ctx).UserByID.Load(ctx, obj.CustomerID)\n}\n```\n\nLoaders are **per request**, installed by middleware. A process-wide loader\ncaches across users and leaks data between them.\n\n```go\nfunc withLoaders(svc UserService, next http.Handler) http.Handler {\n    return http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {\n        ctx := context.WithValue(r.Context(), loadersKey{}, NewLoaders(svc))\n        next.ServeHTTP(w, r.WithContext(ctx))\n    })\n}\n```\n\nThe batch function must return results **in the order of the keys it was\ngiven**, with a nil entry and an error per missing key. Returning a shorter\nslice silently misaligns every result.\n\n## 4. Bound Every Query\n\nA public GraphQL endpoint without limits is a denial-of-service endpoint.\n\n```go\nsrv := handler.New(generated.NewExecutableSchema(cfg))\nsrv.AddTransport(transport.POST{})\nsrv.SetQueryCache(lru.New[*ast.QueryDocument](1000))\nsrv.Use(extension.FixedComplexityLimit(300))\nsrv.Use(extension.AutomaticPersistedQuery{Cache: lru.New[string](100)})\n```\n\n- **Complexity limit** — assign a cost per field, higher for list fields with\n  a large `first`. Start at a number your slowest legitimate query fits under,\n  then measure.\n- **Depth** — recursive types (`user { orders { customer { orders ... } } }`)\n  must be bounded. gqlgen has no built-in depth limit; enforce it in an\n  operation middleware.\n- **Pagination is mandatory** on every list field. A field returning an\n  unbounded list is a schema bug.\n- **Introspection** is only enabled if you install `extension.Introspection`.\n  Do not install it in production, or gate it behind an authenticated role.\n- **Persisted queries** let a public client send a hash instead of a document,\n  so the server executes only queries you shipped.\n\nSet `srv.AroundOperations` to enforce a per-operation timeout, and always run\nbehind an `http.Server` with `ReadTimeout` and `WriteTimeout` set.\n\n## 5. Errors\n\nGraphQL returns 200 with an `errors` array. Never leak internals into it.\n\n```go\nsrv.SetErrorPresenter(func(ctx context.Context, e error) *gqlerror.Error {\n    err := graphql.DefaultErrorPresenter(ctx, e)\n\n    var domainErr *domain.ValidationError\n    if errors.As(e, &domainErr) {\n        err.Message = domainErr.Message\n        err.Extensions = map[string]any{\"code\": \"VALIDATION_FAILED\"}\n        return err\n    }\n\n    slog.ErrorContext(ctx, \"graphql resolver failed\", \"error\", e)\n    err.Message = \"internal server error\" \u002F\u002F stable, safe\n    err.Extensions = map[string]any{\"code\": \"INTERNAL\"}\n    return err\n})\n```\n\nUse `srv.SetRecoverFunc` to convert a resolver panic into an error instead of\nkilling the connection, and log it with the stack.\n\nRemember the nullability rule: an error on a non-null field nulls out its\nnearest nullable ancestor. Make a field non-null only when it can never\nlegitimately be absent.\n\n## 6. Authorization Belongs on the Field\n\nObject-level checks are not enough — a client can reach an object through\nseveral paths.\n\n```graphql\ndirective @hasRole(role: Role!) on FIELD_DEFINITION\n\ntype User {\n  id: ID!\n  email: String! @hasRole(role: ADMIN)\n}\n```\n\n```go\ncfg.Directives.HasRole = func(ctx context.Context, obj any, next graphql.Resolver, role model.Role) (any, error) {\n    if !auth.FromContext(ctx).HasRole(role) {\n        return nil, gqlerror.Errorf(\"access denied\")\n    }\n    return next(ctx)\n}\n```\n\nAuthenticate in HTTP middleware, before the GraphQL handler. Authorize in the\ndirective or the resolver, using the principal from the context — never from\na query argument.\n\n## 7. Testing\n\n```go\nfunc TestUserQuery(t *testing.T) {\n    c := client.New(handler.NewDefaultServer(generated.NewExecutableSchema(cfg)))\n\n    var resp struct {\n        User struct{ ID, Email string }\n    }\n    c.MustPost(`{ user(id: \"u-1\") { id email } }`, &resp)\n\n    require.Equal(t, \"u-1\", resp.User.ID)\n}\n```\n\nAssert the query count for any resolver with a dataloader — that is the only\nway an N+1 regression fails a build rather than a dashboard:\n\n```go\nrequire.Equal(t, 2, db.QueryCount(), \"expected batched loads, got N+1\")\n```\n\n## Verification Checklist\n\n1. Schema is the source of truth; generated files are committed and CI-checked\n2. Generated types bind to domain models via `gqlgen.yml`\n3. Resolvers contain no business logic and always propagate `ctx`\n4. Every list-field resolver that fetches by ID goes through a dataloader\n5. Dataloaders are constructed per request, never shared across requests\n6. Batch functions return one result per key, in key order\n7. A complexity limit and a depth bound are configured and tested\n8. Every list field is paginated\n9. Introspection is disabled or role-gated in production\n10. An error presenter strips internal errors; a recover func is installed\n11. Authorization is enforced per field, from the context principal\n12. A test asserts the query count for at least one batched field\n",{"name":5,"license":15,"metadata":24,"description":27,"allowed-tools":28,"compatibility":29,"user-invocable":30},{"author":25,"version":26},"eduardo-sl","1.0.1","Build GraphQL servers in Go with gqlgen: schema-first generation, resolver structure, the N+1 problem and dataloaders, complexity and depth limits, error presentation, field-level authorization, and testing resolvers. Use when implementing or reviewing a GraphQL API, when a query fans out into hundreds of database calls, or when deciding what a resolver may expose. Trigger examples: \"GraphQL\", \"gqlgen\", \"resolver\", \"N+1 queries\", \"dataloader\", \"query complexity limit\", \"GraphQL schema\". Not for: REST and OpenAPI (go-openapi), gRPC (go-grpc), general HTTP middleware and shutdown (go-api-design).\n","Read Edit Write Glob Grep Bash(go:*) Bash(gofmt:*) Bash(gqlgen:*)","Designed for Claude Code or similar AI coding agents working on Go projects. Requires the Go toolchain. gqlgen is installed as a module tool.",true,"main","19f4d52a98d557f3a645a2cfbbeb5bd54f28831d",".github\u002Fskills\u002Fgo-agent-skills\u002Fskills\u002F(architecture)\u002Fgo-graphql","https:\u002F\u002Fgithub.com\u002Fkongnakornna\u002Fgolangapi","https:\u002F\u002Fgithub.com\u002Fkongnakornna","http:\u002F\u002Fcmoniot.trueddns.com:52160","Technical Skills and Expertise\r\nLanguages and Frameworks\r\nFrontend: HTML, JavaScript, TypeScript, React, Vue, Angular, Next.js\r\nBackend: Node.js,Python,Golang","2026-10-06 23:01:46",{"source":8,"skillName":5,"commands":40,"prompt":50,"zipAvailable":30},[41,44,47],{"tool":42,"command":43},"npx","npx skills add kongnakornna\u002Fgolangapi --skill go-graphql",{"tool":45,"command":46},"bunx","bunx skills add kongnakornna\u002Fgolangapi --skill go-graphql",{"tool":48,"command":49},"pnpm","pnpm dlx skills add kongnakornna\u002Fgolangapi --skill go-graphql","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fkongnakornna\u002Fgolangapi\u002Fblob\u002Fmain\u002F.github\u002Fskills\u002Fgo-agent-skills\u002Fskills\u002F(architecture)\u002Fgo-graphql\u002FSKILL.md",[],[53,61,68,75,82,89],{"skillNo":54,"name":55,"nameZh":6,"description":56,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":57,"licenseFileUrl":6,"sourceHtmlUrl":58,"categories":59,"keywords":60,"publishedAt":19},"01M7A0M9RJJ3383VEFYKABWV15","tb-writing-plans","Use when you have a spec or requirements for a multi-step task, before touching code","Apache-2.0","https:\u002F\u002Fgithub.com\u002Fkongnakornna\u002Fgolangapi\u002Fblob\u002Fmain\u002F.github\u002Fskills\u002Ftb-writing-plans\u002FSKILL.md",[],[],{"skillNo":62,"name":63,"nameZh":6,"description":64,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":57,"licenseFileUrl":6,"sourceHtmlUrl":65,"categories":66,"keywords":67,"publishedAt":19},"01M7A0M4JKY4DEDRP1CBHT00WV","tb-using-git-worktrees","Use when starting feature work that needs isolation from current workspace or before executing implementation plans - ensures an isolated workspace exists via native tools or git worktree fallback","https:\u002F\u002Fgithub.com\u002Fkongnakornna\u002Fgolangapi\u002Fblob\u002Fmain\u002F.github\u002Fskills\u002Ftb-using-git-worktrees\u002FSKILL.md",[],[],{"skillNo":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":57,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"publishedAt":19},"01M7A0KZ5HRAQAAT2KXR6146Q7","tb-test-driven-development","Use when implementing any feature or bugfix, before writing implementation code","https:\u002F\u002Fgithub.com\u002Fkongnakornna\u002Fgolangapi\u002Fblob\u002Fmain\u002F.github\u002Fskills\u002Ftb-test-driven-development\u002FSKILL.md",[],[],{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":57,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"publishedAt":19},"01M7A0KW92W3F18E2MRP8H62P2","tb-subagent-driven-development","Use when executing implementation plans with independent tasks in the current session","https:\u002F\u002Fgithub.com\u002Fkongnakornna\u002Fgolangapi\u002Fblob\u002Fmain\u002F.github\u002Fskills\u002Ftb-subagent-driven-development\u002FSKILL.md",[],[],{"skillNo":83,"name":84,"nameZh":6,"description":85,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":57,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":88,"publishedAt":19},"01M7A0KKYQCHYYRHA5J5BAP24M","tb-scrutinize","Outsider-perspective end-to-end review of a plan, PR, or code change. First questions intent and whether a simpler\u002Fmore elegant approach would achieve the same goal, then traces the actual code path (not just the diff) to verify the change does what it claims. Output is concise, actionable, and every call carries its rationale. Trigger on \u002Ftb-scrutinize and proactively whenever the user asks to review, audit, sanity-check, or get a second opinion on a plan, PR, diff, design doc, or proposed code change.","https:\u002F\u002Fgithub.com\u002Fkongnakornna\u002Fgolangapi\u002Fblob\u002Fmain\u002F.github\u002Fskills\u002Ftb-scrutinize\u002FSKILL.md",[],[],{"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":57,"licenseFileUrl":6,"sourceHtmlUrl":93,"categories":94,"keywords":95,"publishedAt":19},"01M7A0KFDR1V3RW7SW1XNHP82P","tb-requesting-code-review","Use when completing tasks, implementing major features, or before merging to verify work meets requirements","https:\u002F\u002Fgithub.com\u002Fkongnakornna\u002Fgolangapi\u002Fblob\u002Fmain\u002F.github\u002Fskills\u002Ftb-requesting-code-review\u002FSKILL.md",[],[]]