tribunal-kit 4.5.1 → 4.6.1

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (214) hide show
  1. package/.agent/.shared/ui-ux-pro-max/README.md +4 -4
  2. package/.agent/ARCHITECTURE.md +282 -277
  3. package/.agent/agents/accessibility-reviewer.md +187 -187
  4. package/.agent/agents/ai-code-reviewer.md +199 -199
  5. package/.agent/agents/api-architect.md +71 -66
  6. package/.agent/agents/backend-specialist.md +219 -215
  7. package/.agent/agents/cloud-engineer.md +98 -0
  8. package/.agent/agents/code-archaeologist.md +168 -161
  9. package/.agent/agents/database-architect.md +184 -184
  10. package/.agent/agents/db-latency-auditor.md +213 -216
  11. package/.agent/agents/debugger.md +198 -191
  12. package/.agent/agents/dependency-reviewer.md +106 -103
  13. package/.agent/agents/devops-engineer.md +218 -218
  14. package/.agent/agents/documentation-writer.md +209 -201
  15. package/.agent/agents/explorer-agent.md +167 -160
  16. package/.agent/agents/frontend-reviewer.md +162 -160
  17. package/.agent/agents/frontend-specialist.md +257 -248
  18. package/.agent/agents/game-developer.md +48 -48
  19. package/.agent/agents/logic-reviewer.md +118 -116
  20. package/.agent/agents/mobile-developer.md +197 -200
  21. package/.agent/agents/mobile-reviewer.md +159 -162
  22. package/.agent/agents/orchestrator.md +187 -181
  23. package/.agent/agents/penetration-tester.md +160 -157
  24. package/.agent/agents/performance-optimizer.md +183 -183
  25. package/.agent/agents/performance-reviewer.md +178 -178
  26. package/.agent/agents/precedence-reviewer.md +251 -250
  27. package/.agent/agents/product-manager.md +149 -142
  28. package/.agent/agents/product-owner.md +81 -80
  29. package/.agent/agents/project-planner.md +152 -142
  30. package/.agent/agents/qa-automation-engineer.md +216 -225
  31. package/.agent/agents/resilience-reviewer.md +88 -88
  32. package/.agent/agents/schema-reviewer.md +67 -67
  33. package/.agent/agents/security-auditor.md +180 -174
  34. package/.agent/agents/seo-specialist.md +188 -193
  35. package/.agent/agents/sql-reviewer.md +159 -161
  36. package/.agent/agents/supervisor-agent.md +173 -184
  37. package/.agent/agents/swarm-worker-contracts.md +170 -166
  38. package/.agent/agents/swarm-worker-registry.md +92 -92
  39. package/.agent/agents/system-architect.md +85 -0
  40. package/.agent/agents/test-coverage-reviewer.md +158 -160
  41. package/.agent/agents/test-engineer.md +118 -118
  42. package/.agent/agents/throughput-optimizer.md +291 -299
  43. package/.agent/agents/type-safety-reviewer.md +182 -175
  44. package/.agent/agents/ui-ux-auditor.md +300 -292
  45. package/.agent/agents/vitals-reviewer.md +223 -223
  46. package/.agent/mcp_config.json +37 -40
  47. package/.agent/patterns/generator.md +11 -9
  48. package/.agent/patterns/inversion.md +14 -12
  49. package/.agent/patterns/pipeline.md +11 -9
  50. package/.agent/patterns/reviewer.md +15 -13
  51. package/.agent/patterns/tool-wrapper.md +11 -9
  52. package/.agent/routing_index.json +714 -0
  53. package/.agent/rules/GEMINI.md +359 -352
  54. package/.agent/scripts/compile_router.py +112 -0
  55. package/.agent/scripts/migrate_skills_frontmatter.py +64 -0
  56. package/.agent/scripts/strengthen_skills.js +1 -1
  57. package/.agent/skills/advanced-rag-pipelines/SKILL.md +56 -0
  58. package/.agent/skills/agent-organizer/SKILL.md +156 -150
  59. package/.agent/skills/agentic-patterns/SKILL.md +313 -315
  60. package/.agent/skills/ai-prompt-injection-defense/SKILL.md +190 -184
  61. package/.agent/skills/api-patterns/SKILL.md +253 -247
  62. package/.agent/skills/api-security-auditor/SKILL.md +195 -193
  63. package/.agent/skills/app-builder/SKILL.md +573 -572
  64. package/.agent/skills/app-builder/templates/SKILL.md +108 -115
  65. package/.agent/skills/app-builder/templates/astro-static/TEMPLATE.md +76 -76
  66. package/.agent/skills/app-builder/templates/chrome-extension/TEMPLATE.md +92 -92
  67. package/.agent/skills/app-builder/templates/cli-tool/TEMPLATE.md +88 -88
  68. package/.agent/skills/app-builder/templates/electron-desktop/TEMPLATE.md +88 -88
  69. package/.agent/skills/app-builder/templates/express-api/TEMPLATE.md +83 -83
  70. package/.agent/skills/app-builder/templates/flutter-app/TEMPLATE.md +90 -90
  71. package/.agent/skills/app-builder/templates/monorepo-turborepo/TEMPLATE.md +90 -90
  72. package/.agent/skills/app-builder/templates/nextjs-fullstack/TEMPLATE.md +126 -122
  73. package/.agent/skills/app-builder/templates/nextjs-saas/TEMPLATE.md +127 -122
  74. package/.agent/skills/app-builder/templates/nextjs-static/TEMPLATE.md +172 -169
  75. package/.agent/skills/app-builder/templates/nuxt-app/TEMPLATE.md +139 -134
  76. package/.agent/skills/app-builder/templates/python-fastapi/TEMPLATE.md +83 -83
  77. package/.agent/skills/app-builder/templates/react-native-app/TEMPLATE.md +122 -119
  78. package/.agent/skills/appflow-wireframe/SKILL.md +146 -145
  79. package/.agent/skills/architecture/SKILL.md +226 -219
  80. package/.agent/skills/authentication-best-practices/SKILL.md +197 -189
  81. package/.agent/skills/backend-security-expert/SKILL.md +16 -2
  82. package/.agent/skills/bash-linux/SKILL.md +179 -179
  83. package/.agent/skills/behavioral-modes/SKILL.md +239 -223
  84. package/.agent/skills/brainstorming/SKILL.md +498 -486
  85. package/.agent/skills/browser-native-ai/SKILL.md +57 -4
  86. package/.agent/skills/building-native-ui/SKILL.md +202 -202
  87. package/.agent/skills/cicd-pro/SKILL.md +442 -0
  88. package/.agent/skills/clean-code/SKILL.md +400 -381
  89. package/.agent/skills/cloud-architect/SKILL.md +439 -0
  90. package/.agent/skills/code-review-checklist/SKILL.md +203 -194
  91. package/.agent/skills/config-validator/SKILL.md +165 -165
  92. package/.agent/skills/containerization-pro/SKILL.md +452 -0
  93. package/.agent/skills/csharp-developer/SKILL.md +518 -518
  94. package/.agent/skills/data-validation-schemas/SKILL.md +333 -328
  95. package/.agent/skills/database-design/SKILL.md +247 -240
  96. package/.agent/skills/deployment-procedures/SKILL.md +172 -169
  97. package/.agent/skills/devops-engineer/SKILL.md +345 -345
  98. package/.agent/skills/devops-incident-responder/SKILL.md +143 -137
  99. package/.agent/skills/documentation-templates/SKILL.md +291 -279
  100. package/.agent/skills/edge-computing/SKILL.md +183 -181
  101. package/.agent/skills/emil-design-eng/SKILL.md +147 -0
  102. package/.agent/skills/error-resilience/SKILL.md +411 -428
  103. package/.agent/skills/extract-design-system/SKILL.md +160 -158
  104. package/.agent/skills/framer-motion-expert/SKILL.md +253 -244
  105. package/.agent/skills/frontend-design/SKILL.md +208 -201
  106. package/.agent/skills/frontend-security-expert/SKILL.md +16 -3
  107. package/.agent/skills/game-design-expert/SKILL.md +132 -129
  108. package/.agent/skills/game-engineering-expert/SKILL.md +148 -146
  109. package/.agent/skills/generative-ui-expert/SKILL.md +57 -1
  110. package/.agent/skills/geo-fundamentals/SKILL.md +148 -147
  111. package/.agent/skills/git-pro/SKILL.md +435 -0
  112. package/.agent/skills/github-operations/SKILL.md +335 -329
  113. package/.agent/skills/gsap-core/SKILL.md +319 -308
  114. package/.agent/skills/gsap-frameworks/SKILL.md +213 -207
  115. package/.agent/skills/gsap-performance/SKILL.md +139 -133
  116. package/.agent/skills/gsap-plugins/SKILL.md +486 -480
  117. package/.agent/skills/gsap-react/SKILL.md +202 -189
  118. package/.agent/skills/gsap-scrolltrigger/SKILL.md +357 -350
  119. package/.agent/skills/gsap-timeline/SKILL.md +165 -161
  120. package/.agent/skills/gsap-utils/SKILL.md +344 -338
  121. package/.agent/skills/harness-protocol/SKILL.md +48 -0
  122. package/.agent/skills/i18n-localization/SKILL.md +174 -163
  123. package/.agent/skills/intelligent-routing/SKILL.md +202 -246
  124. package/.agent/skills/knowledge-graph/SKILL.md +60 -52
  125. package/.agent/skills/lint-and-validate/SKILL.md +261 -261
  126. package/.agent/skills/llm-engineering/SKILL.md +400 -394
  127. package/.agent/skills/local-first/SKILL.md +178 -178
  128. package/.agent/skills/mcp-builder/SKILL.md +143 -142
  129. package/.agent/skills/mobile-design/SKILL.md +272 -263
  130. package/.agent/skills/monorepo-management/SKILL.md +335 -334
  131. package/.agent/skills/motion-engineering/SKILL.md +266 -234
  132. package/.agent/skills/nextjs-react-expert/SKILL.md +236 -234
  133. package/.agent/skills/nodejs-best-practices/SKILL.md +547 -548
  134. package/.agent/skills/observability/SKILL.md +343 -343
  135. package/.agent/skills/parallel-agents/SKILL.md +143 -146
  136. package/.agent/skills/performance-profiling/SKILL.md +259 -267
  137. package/.agent/skills/plan-writing/SKILL.md +150 -142
  138. package/.agent/skills/platform-engineer/SKILL.md +148 -147
  139. package/.agent/skills/playwright-best-practices/SKILL.md +188 -187
  140. package/.agent/skills/powershell-windows/SKILL.md +162 -162
  141. package/.agent/skills/project-idioms/SKILL.md +137 -137
  142. package/.agent/skills/python-patterns/SKILL.md +260 -259
  143. package/.agent/skills/python-pro/SKILL.md +324 -323
  144. package/.agent/skills/react-specialist/SKILL.md +305 -277
  145. package/.agent/skills/readme-builder/SKILL.md +310 -300
  146. package/.agent/skills/realtime-patterns/SKILL.md +323 -319
  147. package/.agent/skills/red-team-tactics/SKILL.md +231 -218
  148. package/.agent/skills/review-animations/SKILL.md +72 -0
  149. package/.agent/skills/review-animations/STANDARDS.md +73 -0
  150. package/.agent/skills/rust-pro/SKILL.md +671 -673
  151. package/.agent/skills/seo-fundamentals/SKILL.md +179 -179
  152. package/.agent/skills/server-management/SKILL.md +218 -214
  153. package/.agent/skills/shadcn-ui-expert/SKILL.md +231 -231
  154. package/.agent/skills/skill-creator/SKILL.md +87 -86
  155. package/.agent/skills/sql-pro/SKILL.md +629 -629
  156. package/.agent/skills/supabase-postgres-best-practices/SKILL.md +97 -97
  157. package/.agent/skills/swiftui-expert/SKILL.md +204 -201
  158. package/.agent/skills/system-design-pro/SKILL.md +345 -0
  159. package/.agent/skills/systematic-debugging/SKILL.md +153 -142
  160. package/.agent/skills/tailwind-patterns/SKILL.md +610 -566
  161. package/.agent/skills/tdd-workflow/SKILL.md +169 -161
  162. package/.agent/skills/test-result-analyzer/SKILL.md +313 -309
  163. package/.agent/skills/testing-patterns/SKILL.md +566 -579
  164. package/.agent/skills/trend-researcher/SKILL.md +243 -237
  165. package/.agent/skills/typescript-advanced/SKILL.md +336 -335
  166. package/.agent/skills/ui-ux-pro-max/SKILL.md +590 -562
  167. package/.agent/skills/ui-ux-researcher/SKILL.md +244 -244
  168. package/.agent/skills/vue-expert/SKILL.md +294 -275
  169. package/.agent/skills/vulnerability-scanner/SKILL.md +416 -404
  170. package/.agent/skills/web-accessibility-auditor/SKILL.md +219 -218
  171. package/.agent/skills/web-design-guidelines/SKILL.md +192 -186
  172. package/.agent/skills/webapp-testing/SKILL.md +167 -169
  173. package/.agent/skills/webgpu-performance/SKILL.md +56 -2
  174. package/.agent/skills/whimsy-injector/SKILL.md +346 -325
  175. package/.agent/skills/workflow-optimizer/SKILL.md +231 -229
  176. package/.agent/workflows/acf.md +141 -0
  177. package/.agent/workflows/api-tester.md +176 -151
  178. package/.agent/workflows/audit.md +150 -127
  179. package/.agent/workflows/brainstorm.md +134 -110
  180. package/.agent/workflows/changelog.md +140 -112
  181. package/.agent/workflows/create.md +168 -124
  182. package/.agent/workflows/debug.md +190 -165
  183. package/.agent/workflows/deploy.md +201 -180
  184. package/.agent/workflows/enhance.md +154 -128
  185. package/.agent/workflows/fix.md +136 -114
  186. package/.agent/workflows/generate.md +198 -183
  187. package/.agent/workflows/marathon.md +37 -11
  188. package/.agent/workflows/migrate.md +184 -160
  189. package/.agent/workflows/orchestrate.md +192 -168
  190. package/.agent/workflows/performance-benchmarker.md +135 -114
  191. package/.agent/workflows/plan.md +196 -173
  192. package/.agent/workflows/preview.md +103 -80
  193. package/.agent/workflows/refactor.md +192 -161
  194. package/.agent/workflows/review-ai.md +125 -101
  195. package/.agent/workflows/review.md +141 -116
  196. package/.agent/workflows/session.md +122 -94
  197. package/.agent/workflows/status.md +101 -79
  198. package/.agent/workflows/strengthen-skills.md +164 -138
  199. package/.agent/workflows/super-prompt.md +24 -0
  200. package/.agent/workflows/swarm.md +193 -179
  201. package/.agent/workflows/test.md +211 -189
  202. package/.agent/workflows/tribunal-backend.md +136 -105
  203. package/.agent/workflows/tribunal-database.md +129 -95
  204. package/.agent/workflows/tribunal-frontend.md +140 -96
  205. package/.agent/workflows/tribunal-full.md +131 -100
  206. package/.agent/workflows/tribunal-mobile.md +129 -95
  207. package/.agent/workflows/tribunal-performance.md +136 -110
  208. package/.agent/workflows/tribunal-speed.md +209 -183
  209. package/.agent/workflows/ui-ux-pro-max.md +155 -122
  210. package/README.md +107 -55
  211. package/mcp_config.json +1 -3
  212. package/package.json +94 -94
  213. package/.agent/GEMINI.md +0 -121
  214. package/.agent/skills/doc.md +0 -177
@@ -1,251 +1,255 @@
1
- ---
2
- name: api-patterns
3
- description: API design mastery. REST, GraphQL, tRPC, and gRPC selection. Request/response design, pagination (cursor/offset), filtering, versioning, rate limiting, error formats (RFC 9457), authentication (JWT/OAuth2/API keys), idempotency, file uploads, webhooks, and OpenAPI documentation. Use when designing APIs, choosing protocols, or implementing API standards.
4
- allowed-tools: Read, Write, Edit, Glob, Grep
5
- version: 3.1.0
6
- last-updated: 2026-04-07
7
- applies-to-model: gemini-3-1-pro, claude-3-7-sonnet
8
- ---
9
-
10
- # API Patterns — Design & Protocol Mastery
11
-
12
- ## Hallucination Traps (Read First)
13
- - JWT in URL query params → ✅ `Authorization: Bearer` header only. Query params get logged in server access logs.
14
- - ❌ Assuming JWT is encrypted → ✅ JWT is base64-encoded (NOT encrypted). Anyone can decode it. Never put secrets/PII in the payload.
15
- - Offset pagination on large tables → ✅ `OFFSET 100000` scans and discards 100K rows. Use cursor pagination for tables > 10K rows.
16
- - ❌ Verbs in REST URLs (`/api/getUsers`) → ✅ Nouns only (`GET /api/users`). HTTP method IS the verb.
17
- - ❌ `POST` is idempotent → ✅ `POST` is NOT idempotent requires `Idempotency-Key` header for safe retries.
18
- - ❌ GraphQL has no security risks → ✅ Deeply nested queries are a DoS vector. Set max depth, query cost limits. Disable introspection in production.
19
-
20
- ---
21
-
22
- ## Protocol Selection Matrix
23
-
24
- | Protocol | Use When |
25
- |----------|----------|
26
- | **REST** | Public APIs, 3rd-party consumers, standard CRUD, HTTP caching |
27
- | **GraphQL** | Complex nested data, multiple clients, flexible queries, mobile bandwidth sensitivity |
28
- | **tRPC** | Full-stack TypeScript (Next.js monorepo), shared types, no codegen |
29
- | **gRPC** | Internal microservices, high-throughput, streaming, binary protocol |
30
- | **WebSocket** | Bidirectional real-time (chat, gaming, live collaboration) |
31
- | **SSE** | Server-to-client streaming only (AI token streaming, live feeds) |
32
-
33
- ---
34
-
35
- ## REST Design
36
-
37
- ### URL Conventions
38
- ```
39
- ✅ GET /api/v1/users list users
40
- ✅ GET /api/v1/users/123 get user by ID
41
- ✅ POST /api/v1/users create user
42
- ✅ PATCH /api/v1/users/123 partial update
43
- ✅ DELETE /api/v1/users/123 delete user
44
- ✅ GET /api/v1/users/123/posts nested resource
45
-
46
- /api/getUsers /api/createUser /api/user (singular) /api/Users (uppercase)
47
- ```
48
-
49
- ### HTTP Status Codes
50
- ```
51
- 200 OK → GET / PUT / PATCH success
52
- 201 Created → POST success (include Location: /api/v1/users/123 header)
53
- 204 No Content → DELETE success
54
- 400 Bad Request → Malformed request / missing fields
55
- 401 Unauthorized → Missing or invalid authentication
56
- 403 Forbidden → Authenticated but not authorized
57
- 404 Not Found Resource does not exist
58
- 409 Conflict Duplicate resource (email already exists)
59
- 422 Unprocessable Valid JSON, semantically invalid data
60
- 429 Too Many Req Rate limit exceeded
61
- 500 Internal Unhandled server error — NEVER expose stack traces
62
- ```
63
-
64
- ### Response Envelope
65
- ```typescript
66
- interface ApiResponse<T> { data: T; meta?: Record<string, unknown>; }
67
-
68
- interface ApiError {
69
- error: {
70
- code: string; // machine-readable: "VALIDATION_ERROR"
71
- message: string; // human-readable: "Email is already in use"
72
- details?: Array<{ field: string; message: string }>; // field-level errors
73
- requestId?: string; // for support/tracing
74
- };
75
- }
76
- ```
77
-
78
- ---
79
-
80
- ## Pagination
81
-
82
- ```typescript
83
- // Cursor-based — required for large/dynamic datasets
84
- // GET /api/v1/posts?cursor=eyJpZCI6MTAwfQ&limit=20
85
- const posts = await db.post.findMany({
86
- where: { id: { lt: decodeCursor(req.query.cursor).id } },
87
- orderBy: { id: "desc" },
88
- take: limit + 1, // fetch one extra to determine hasMore
89
- });
90
- const hasMore = posts.length > limit;
91
- if (hasMore) posts.pop();
92
- return { data: posts, meta: { hasMore, nextCursor: encodeCursor(posts.at(-1)) } };
93
-
94
- // Offset-based — only for small datasets where users need page jumping
95
- // GET /api/v1/posts?page=3&limit=20
96
- // TRAP: OFFSET 100000 scans and discards 100K rows — degrades badly at scale
97
- ```
98
-
99
- ---
100
-
101
- ## Idempotency
102
-
103
- ```typescript
104
- // POST /api/v1/payments with header: Idempotency-Key: <uuid>
105
- app.post("/api/v1/payments", async (req, res) => {
106
- const key = req.headers["idempotency-key"];
107
- if (!key) return res.status(400).json({ error: "Missing Idempotency-Key" });
108
-
109
- const cached = await redis.get(`idempotency:${key}`);
110
- if (cached) return res.status(200).json(JSON.parse(cached));
111
-
112
- const result = await processPayment(req.body);
113
- await redis.set(`idempotency:${key}`, JSON.stringify(result), "EX", 86400);
114
- return res.status(201).json(result);
115
- });
116
- // GET, PUT, DELETE → naturally idempotent (safe to retry without a key)
117
- // POST, PATCH → NOT idempotent by default — require Idempotency-Key
118
- ```
119
-
120
- ---
121
-
122
- ## Webhooks
123
-
124
- ```typescript
125
- // HMAC signature verification (always verify — never trust unsigned webhooks)
126
- import { createHmac, timingSafeEqual } from "node:crypto";
127
- function verify(payload: string, signature: string, secret: string): boolean {
128
- const expected = createHmac("sha256", secret).update(payload).digest("hex");
129
- return timingSafeEqual(Buffer.from(signature), Buffer.from(expected));
130
- }
131
-
132
- app.post("/webhooks", (req, res) => {
133
- if (!verify(JSON.stringify(req.body), req.headers["x-webhook-signature"] as string, WEBHOOK_SECRET))
134
- return res.status(401).send("Invalid signature");
135
- res.status(200).send("OK"); // respond immediately
136
- processWebhookAsync(req.body); // process asynchronously
137
- });
138
- // Retry policy: 3 retries with exponential backoff (1s → 10s → 100s)
139
- // Include unique event ID in payload for receiver-side deduplication
140
- ```
141
-
142
- ---
143
-
144
- ## Versioning
145
-
146
- ```
147
- URL path (recommended): /api/v1/users simplest, most common, cache-friendly
148
- Header: Accept: application/vnd.api.v1+json
149
- Query param: /api/users?version=1 → messy, avoid
150
-
151
- Rules:
152
- - Start at v1, never v0
153
- - Breaking changes = new major version (v2)
154
- - Non-breaking additions (new optional fields) do NOT need a version bump
155
- - Deprecate before removing — give consumers 6+ months notice
156
- ```
157
-
158
- ---
159
-
160
- ## Rate Limiting
161
-
162
- ```
163
- Strategy How When
164
- Token bucket → Burst allowed, refills Most APIs (recommended)
165
- Sliding window → Smooth distribution Strict fairness required
166
- Fixed window → Simple counter per period Basic needs only
167
-
168
- Response headers to always include:
169
- X-RateLimit-Limit (max requests in window)
170
- X-RateLimit-Remaining (requests left)
171
- X-RateLimit-Reset (Unix timestamp when limit resets)
172
- Retry-After (seconds to wait on 429)
173
- ```
174
-
175
- ---
176
-
177
- ## GraphQL Security
178
-
179
- ```
180
- Protect against:
181
- Depth attacks → Set max query depth (typically 7–10)
182
- Cost attacks → Calculate query complexity score, reject > threshold
183
- Batch abuse → Limit batch size / alias count
184
- Introspection → Disable in production (exposes full schema to attackers)
185
- ```
186
-
187
- ---
188
-
189
- ## Authentication Selection
190
-
191
- | Pattern | Best For |
192
- |---------|----------|
193
- | **JWT** (short-lived access + httpOnly refresh) | Stateless services, microservices |
194
- | **Session** | Traditional server-rendered apps |
195
- | **OAuth 2.0 / OIDC** | Third-party login, delegated access |
196
- | **API Key** | Server-to-server, public API consumers |
197
- | **Passkey (WebAuthn)** | Modern passwordless (2026+) |
198
-
199
-
200
- ---
201
-
202
-
203
-
204
- AI coding assistants often fall into specific bad habits when dealing with this domain. These are strictly forbidden:
205
-
206
- 1. **Over-engineering:** Proposing complex abstractions or distributed systems when a simpler approach suffices.
207
- 2. **Hallucinated Libraries/Methods:** Using non-existent methods or packages. Always `// VERIFY` or check `package.json` / `requirements.txt`.
208
- 3. **Skipping Edge Cases:** Writing the "happy path" and ignoring error handling, timeouts, or data validation.
209
- 4. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
210
- 5. **Silent Degradation:** Catching and suppressing errors without logging or re-raising.
211
-
212
- ---
213
-
214
-
215
-
216
- **Slash command: `/review` or `/tribunal-full`**
217
- **Active reviewers: `logic-reviewer` · `security-auditor`**
218
-
219
- ### ❌ Forbidden AI Tropes
220
-
221
- 1. **Blind Assumptions:** Never make an assumption without documenting it clearly with `// VERIFY: [reason]`.
222
- 2. **Silent Degradation:** Catching and suppressing errors without logging or handling.
223
- 3. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
224
-
225
-
226
-
227
- Review these questions before confirming output:
228
- ```
229
- ✅ Did I rely ONLY on real, verified tools and methods?
230
- ✅ Is this solution appropriately scoped to the user's constraints?
231
- ✅ Did I handle potential failure modes and edge cases?
232
- ✅ Have I avoided generic boilerplate that doesn't add value?
233
- ```
234
-
235
- ### 🛑 Verification-Before-Completion (VBC) Protocol
236
-
237
- **CRITICAL:** You must follow a strict "evidence-based closeout" state machine.
238
- - ❌ **Forbidden:** Declaring a task complete because the output "looks correct."
239
- - ✅ **Required:** You are explicitly forbidden from finalizing any task without providing **concrete evidence** (terminal output, passing tests, compile success, or equivalent proof) that your output works as intended.
240
-
241
-
242
- ## Pre-Flight Checklist
243
- - [ ] Have I reviewed the user's specific constraints and requests?
244
- - [ ] Have I checked the environment for relevant existing implementations?
245
-
246
- ## VBC Protocol (Verification-Before-Completion)
247
- You MUST verify existing code signatures and variables before attempting to modify or call them. No hallucination is permitted.
1
+ ---
2
+ name: api-patterns
3
+ description: API design mastery. REST, GraphQL, tRPC, and gRPC selection. Request/response design, pagination (cursor/offset), filtering, versioning, rate limiting, error formats (RFC 9457), authentication (JWT/OAuth2/API keys), idempotency, file uploads, webhooks, and OpenAPI documentation. Use when designing APIs, choosing protocols, or implementing API standards.
4
+ allowed-tools: Read, Write, Edit, Glob, Grep
5
+ version: 3.1.0
6
+ last-updated: 2026-04-07
7
+ applies-to-model: gemini-3-1-pro, claude-3-7-sonnet
8
+ routing:
9
+ domain: general
10
+ tier: basic
11
+ ---
12
+
13
+ # API Patterns Design & Protocol Mastery
14
+
15
+ ## Hallucination Traps (Read First)
16
+
17
+ - ❌ JWT in URL query params → ✅ `Authorization: Bearer` header only. Query params get logged in server access logs.
18
+ - ❌ Assuming JWT is encrypted → ✅ JWT is base64-encoded (NOT encrypted). Anyone can decode it. Never put secrets/PII in the payload.
19
+ - ❌ Offset pagination on large tables → ✅ `OFFSET 100000` scans and discards 100K rows. Use cursor pagination for tables > 10K rows.
20
+ - ❌ Verbs in REST URLs (`/api/getUsers`) → ✅ Nouns only (`GET /api/users`). HTTP method IS the verb.
21
+ - ❌ `POST` is idempotent → ✅ `POST` is NOT idempotent — requires `Idempotency-Key` header for safe retries.
22
+ - GraphQL has no security risks → ✅ Deeply nested queries are a DoS vector. Set max depth, query cost limits. Disable introspection in production.
23
+
24
+ ---
25
+
26
+ ## Protocol Selection Matrix
27
+
28
+ | Protocol | Use When |
29
+ | ------------- | ------------------------------------------------------------------------------------- |
30
+ | **REST** | Public APIs, 3rd-party consumers, standard CRUD, HTTP caching |
31
+ | **GraphQL** | Complex nested data, multiple clients, flexible queries, mobile bandwidth sensitivity |
32
+ | **tRPC** | Full-stack TypeScript (Next.js monorepo), shared types, no codegen |
33
+ | **gRPC** | Internal microservices, high-throughput, streaming, binary protocol |
34
+ | **WebSocket** | Bidirectional real-time (chat, gaming, live collaboration) |
35
+ | **SSE** | Server-to-client streaming only (AI token streaming, live feeds) |
36
+
37
+ ---
38
+
39
+ ## REST Design
40
+
41
+ ### URL Conventions
42
+
43
+ ```
44
+ ✅ GET /api/v1/users list users
45
+ ✅ GET /api/v1/users/123 get user by ID
46
+ POST /api/v1/users create user
47
+ ✅ PATCH /api/v1/users/123 partial update
48
+ ✅ DELETE /api/v1/users/123 delete user
49
+ ✅ GET /api/v1/users/123/posts nested resource
50
+
51
+ /api/getUsers /api/createUser /api/user (singular) /api/Users (uppercase)
52
+ ```
53
+
54
+ ### HTTP Status Codes
55
+
56
+ ```
57
+ 200 OK GET / PUT / PATCH success
58
+ 201 Created POST success (include Location: /api/v1/users/123 header)
59
+ 204 No Content DELETE success
60
+ 400 Bad Request Malformed request / missing fields
61
+ 401 Unauthorized Missing or invalid authentication
62
+ 403 Forbidden → Authenticated but not authorized
63
+ 404 Not Found → Resource does not exist
64
+ 409 Conflict → Duplicate resource (email already exists)
65
+ 422 Unprocessable → Valid JSON, semantically invalid data
66
+ 429 Too Many Req → Rate limit exceeded
67
+ 500 Internal → Unhandled server error — NEVER expose stack traces
68
+ ```
69
+
70
+ ### Response Envelope
71
+
72
+ ```typescript
73
+ interface ApiResponse<T> {
74
+ data: T;
75
+ meta?: Record<string, unknown>;
76
+ }
77
+
78
+ interface ApiError {
79
+ error: {
80
+ code: string; // machine-readable: "VALIDATION_ERROR"
81
+ message: string; // human-readable: "Email is already in use"
82
+ details?: Array<{ field: string; message: string }>; // field-level errors
83
+ requestId?: string; // for support/tracing
84
+ };
85
+ }
86
+ ```
87
+
88
+ ---
89
+
90
+ ## Pagination
91
+
92
+ ```typescript
93
+ // ✅ Cursor-based — required for large/dynamic datasets
94
+ // GET /api/v1/posts?cursor=eyJpZCI6MTAwfQ&limit=20
95
+ const posts = await db.post.findMany({
96
+ where: { id: { lt: decodeCursor(req.query.cursor).id } },
97
+ orderBy: { id: "desc" },
98
+ take: limit + 1, // fetch one extra to determine hasMore
99
+ });
100
+ const hasMore = posts.length > limit;
101
+ if (hasMore) posts.pop();
102
+ return { data: posts, meta: { hasMore, nextCursor: encodeCursor(posts.at(-1)) } };
103
+
104
+ // Offset-based only for small datasets where users need page jumping
105
+ // GET /api/v1/posts?page=3&limit=20
106
+ // TRAP: OFFSET 100000 scans and discards 100K rows — degrades badly at scale
107
+ ```
108
+
109
+ ---
110
+
111
+ ## Idempotency
112
+
113
+ ```typescript
114
+ // POST /api/v1/payments with header: Idempotency-Key: <uuid>
115
+ app.post("/api/v1/payments", async (req, res) => {
116
+ const key = req.headers["idempotency-key"];
117
+ if (!key) return res.status(400).json({ error: "Missing Idempotency-Key" });
118
+
119
+ const cached = await redis.get(`idempotency:${key}`);
120
+ if (cached) return res.status(200).json(JSON.parse(cached));
121
+
122
+ const result = await processPayment(req.body);
123
+ await redis.set(`idempotency:${key}`, JSON.stringify(result), "EX", 86400);
124
+ return res.status(201).json(result);
125
+ });
126
+ // GET, PUT, DELETE naturally idempotent (safe to retry without a key)
127
+ // POST, PATCH → NOT idempotent by default — require Idempotency-Key
128
+ ```
129
+
130
+ ---
131
+
132
+ ## Webhooks
133
+
134
+ ```typescript
135
+ // HMAC signature verification (always verify never trust unsigned webhooks)
136
+ import { createHmac, timingSafeEqual } from "node:crypto";
137
+ function verify(payload: string, signature: string, secret: string): boolean {
138
+ const expected = createHmac("sha256", secret).update(payload).digest("hex");
139
+ return timingSafeEqual(Buffer.from(signature), Buffer.from(expected));
140
+ }
141
+
142
+ app.post("/webhooks", (req, res) => {
143
+ if (!verify(JSON.stringify(req.body), req.headers["x-webhook-signature"] as string, WEBHOOK_SECRET)) return res.status(401).send("Invalid signature");
144
+ res.status(200).send("OK"); // respond immediately
145
+ processWebhookAsync(req.body); // process asynchronously
146
+ });
147
+ // Retry policy: 3 retries with exponential backoff (1s 10s 100s)
148
+ // Include unique event ID in payload for receiver-side deduplication
149
+ ```
150
+
151
+ ---
152
+
153
+ ## Versioning
154
+
155
+ ```
156
+ URL path (recommended): /api/v1/users → simplest, most common, cache-friendly
157
+ Header: Accept: application/vnd.api.v1+json
158
+ Query param: /api/users?version=1 → messy, avoid
159
+
160
+ Rules:
161
+ - Start at v1, never v0
162
+ - Breaking changes = new major version (v2)
163
+ - Non-breaking additions (new optional fields) do NOT need a version bump
164
+ - Deprecate before removing give consumers 6+ months notice
165
+ ```
166
+
167
+ ---
168
+
169
+ ## Rate Limiting
170
+
171
+ ```
172
+ Strategy How When
173
+ Token bucket → Burst allowed, refills Most APIs (recommended)
174
+ Sliding window → Smooth distribution Strict fairness required
175
+ Fixed window → Simple counter per period Basic needs only
176
+
177
+ Response headers to always include:
178
+ X-RateLimit-Limit (max requests in window)
179
+ X-RateLimit-Remaining (requests left)
180
+ X-RateLimit-Reset (Unix timestamp when limit resets)
181
+ Retry-After (seconds to wait on 429)
182
+ ```
183
+
184
+ ---
185
+
186
+ ## GraphQL Security
187
+
188
+ ```
189
+ Protect against:
190
+ Depth attacks → Set max query depth (typically 7–10)
191
+ Cost attacks → Calculate query complexity score, reject > threshold
192
+ Batch abuse → Limit batch size / alias count
193
+ Introspection → Disable in production (exposes full schema to attackers)
194
+ ```
195
+
196
+ ---
197
+
198
+ ## Authentication Selection
199
+
200
+ | Pattern | Best For |
201
+ | ----------------------------------------------- | -------------------------------------- |
202
+ | **JWT** (short-lived access + httpOnly refresh) | Stateless services, microservices |
203
+ | **Session** | Traditional server-rendered apps |
204
+ | **OAuth 2.0 / OIDC** | Third-party login, delegated access |
205
+ | **API Key** | Server-to-server, public API consumers |
206
+ | **Passkey (WebAuthn)** | Modern passwordless (2026+) |
207
+
208
+ ---
209
+
210
+ AI coding assistants often fall into specific bad habits when dealing with this domain. These are strictly forbidden:
248
211
 
212
+ 1. **Over-engineering:** Proposing complex abstractions or distributed systems when a simpler approach suffices.
213
+ 2. **Hallucinated Libraries/Methods:** Using non-existent methods or packages. Always `// VERIFY` or check `package.json` / `requirements.txt`.
214
+ 3. **Skipping Edge Cases:** Writing the "happy path" and ignoring error handling, timeouts, or data validation.
215
+ 4. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
216
+ 5. **Silent Degradation:** Catching and suppressing errors without logging or re-raising.
217
+
218
+ ---
219
+
220
+ **Slash command: `/review` or `/tribunal-full`**
221
+ **Active reviewers: `logic-reviewer` · `security-auditor`**
222
+
223
+ ### ❌ Forbidden AI Tropes
224
+
225
+ 1. **Blind Assumptions:** Never make an assumption without documenting it clearly with `// VERIFY: [reason]`.
226
+ 2. **Silent Degradation:** Catching and suppressing errors without logging or handling.
227
+ 3. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
228
+
229
+ Review these questions before confirming output:
230
+
231
+ ```
232
+ ✅ Did I rely ONLY on real, verified tools and methods?
233
+ ✅ Is this solution appropriately scoped to the user's constraints?
234
+ ✅ Did I handle potential failure modes and edge cases?
235
+ ✅ Have I avoided generic boilerplate that doesn't add value?
236
+ ```
237
+
238
+ ### 🛑 Verification-Before-Completion (VBC) Protocol
239
+
240
+ **CRITICAL:** You must follow a strict "evidence-based closeout" state machine.
241
+
242
+ - ❌ **Forbidden:** Declaring a task complete because the output "looks correct."
243
+ - ✅ **Required:** You are explicitly forbidden from finalizing any task without providing **concrete evidence** (terminal output, passing tests, compile success, or equivalent proof) that your output works as intended.
244
+
245
+ ## Pre-Flight Checklist
246
+
247
+ - [ ] Have I reviewed the user's specific constraints and requests?
248
+ - [ ] Have I checked the environment for relevant existing implementations?
249
+
250
+ ## VBC Protocol (Verification-Before-Completion)
251
+
252
+ You MUST verify existing code signatures and variables before attempting to modify or call them. No hallucination is permitted.
249
253
 
250
254
  ---
251
255
 
@@ -275,6 +279,7 @@ AI coding assistants often fall into specific bad habits when dealing with this
275
279
  ### ✅ Pre-Flight Self-Audit
276
280
 
277
281
  Review these questions before confirming output:
282
+
278
283
  ```
279
284
  ✅ Did I rely ONLY on real, verified tools and methods?
280
285
  ✅ Is this solution appropriately scoped to the user's constraints?
@@ -285,5 +290,6 @@ Review these questions before confirming output:
285
290
  ### 🛑 Verification-Before-Completion (VBC) Protocol
286
291
 
287
292
  **CRITICAL:** You must follow a strict "evidence-based closeout" state machine.
293
+
288
294
  - ❌ **Forbidden:** Declaring a task complete because the output "looks correct."
289
295
  - ✅ **Required:** You are explicitly forbidden from finalizing any task without providing **concrete evidence** (terminal output, passing tests, compile success, or equivalent proof) that your output works as intended.