tribunal-kit 4.2.0 → 4.3.1

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 (186) hide show
  1. package/.agent/ARCHITECTURE.md +21 -14
  2. package/.agent/agents/swarm-worker-contracts.md +5 -5
  3. package/.agent/agents/ui-ux-auditor.md +292 -0
  4. package/.agent/rules/GEMINI.md +8 -8
  5. package/.agent/scripts/__pycache__/_colors.cpython-311.pyc +0 -0
  6. package/.agent/scripts/__pycache__/_utils.cpython-311.pyc +0 -0
  7. package/.agent/scripts/__pycache__/case_law_manager.cpython-311.pyc +0 -0
  8. package/.agent/scripts/_colors.js +18 -0
  9. package/.agent/scripts/_utils.js +42 -0
  10. package/.agent/scripts/auto_preview.js +197 -0
  11. package/.agent/scripts/bundle_analyzer.js +290 -0
  12. package/.agent/scripts/case_law_manager.js +684 -0
  13. package/.agent/scripts/checklist.js +266 -0
  14. package/.agent/scripts/colors.js +17 -0
  15. package/.agent/scripts/compress_skills.js +141 -0
  16. package/.agent/scripts/consolidate_skills.js +149 -0
  17. package/.agent/scripts/context_broker.js +609 -0
  18. package/.agent/scripts/deep_compress.js +150 -0
  19. package/.agent/scripts/dependency_analyzer.js +272 -0
  20. package/.agent/scripts/graph_builder.js +199 -0
  21. package/.agent/scripts/graph_zoom.js +154 -0
  22. package/.agent/scripts/inner_loop_validator.js +465 -0
  23. package/.agent/scripts/lint_runner.js +187 -0
  24. package/.agent/scripts/minify_context.js +100 -0
  25. package/.agent/scripts/patch_skills_meta.js +156 -0
  26. package/.agent/scripts/patch_skills_output.js +244 -0
  27. package/.agent/scripts/schema_validator.js +297 -0
  28. package/.agent/scripts/security_scan.js +303 -0
  29. package/.agent/scripts/session_manager.js +276 -0
  30. package/.agent/scripts/skill_evolution.js +644 -0
  31. package/.agent/scripts/skill_integrator.js +313 -0
  32. package/.agent/scripts/strengthen_skills.js +193 -0
  33. package/.agent/scripts/strip_tribunal.js +47 -0
  34. package/.agent/scripts/swarm_dispatcher.js +360 -0
  35. package/.agent/scripts/test_runner.js +193 -0
  36. package/.agent/scripts/utils.js +32 -0
  37. package/.agent/scripts/verify_all.js +256 -0
  38. package/.agent/skills/agent-organizer/SKILL.md +12 -4
  39. package/.agent/skills/agentic-patterns/SKILL.md +12 -4
  40. package/.agent/skills/ai-prompt-injection-defense/SKILL.md +12 -4
  41. package/.agent/skills/api-patterns/SKILL.md +209 -201
  42. package/.agent/skills/api-security-auditor/SKILL.md +12 -4
  43. package/.agent/skills/app-builder/SKILL.md +12 -4
  44. package/.agent/skills/app-builder/templates/SKILL.md +76 -68
  45. package/.agent/skills/app-builder/templates/astro-static/TEMPLATE.md +1 -1
  46. package/.agent/skills/app-builder/templates/chrome-extension/TEMPLATE.md +1 -1
  47. package/.agent/skills/app-builder/templates/cli-tool/TEMPLATE.md +1 -1
  48. package/.agent/skills/app-builder/templates/electron-desktop/TEMPLATE.md +1 -1
  49. package/.agent/skills/app-builder/templates/express-api/TEMPLATE.md +1 -1
  50. package/.agent/skills/app-builder/templates/flutter-app/TEMPLATE.md +1 -1
  51. package/.agent/skills/app-builder/templates/monorepo-turborepo/TEMPLATE.md +1 -1
  52. package/.agent/skills/app-builder/templates/nextjs-fullstack/TEMPLATE.md +1 -1
  53. package/.agent/skills/app-builder/templates/nextjs-saas/TEMPLATE.md +1 -1
  54. package/.agent/skills/app-builder/templates/nextjs-static/TEMPLATE.md +1 -1
  55. package/.agent/skills/app-builder/templates/nuxt-app/TEMPLATE.md +1 -1
  56. package/.agent/skills/app-builder/templates/python-fastapi/TEMPLATE.md +1 -1
  57. package/.agent/skills/app-builder/templates/react-native-app/TEMPLATE.md +1 -1
  58. package/.agent/skills/appflow-wireframe/SKILL.md +12 -4
  59. package/.agent/skills/architecture/SKILL.md +12 -4
  60. package/.agent/skills/authentication-best-practices/SKILL.md +12 -4
  61. package/.agent/skills/bash-linux/SKILL.md +12 -4
  62. package/.agent/skills/behavioral-modes/SKILL.md +12 -4
  63. package/.agent/skills/brainstorming/SKILL.md +12 -4
  64. package/.agent/skills/building-native-ui/SKILL.md +12 -4
  65. package/.agent/skills/clean-code/SKILL.md +12 -4
  66. package/.agent/skills/code-review-checklist/SKILL.md +12 -4
  67. package/.agent/skills/config-validator/SKILL.md +12 -4
  68. package/.agent/skills/csharp-developer/SKILL.md +12 -4
  69. package/.agent/skills/data-validation-schemas/SKILL.md +290 -282
  70. package/.agent/skills/database-design/SKILL.md +202 -194
  71. package/.agent/skills/deployment-procedures/SKILL.md +12 -4
  72. package/.agent/skills/devops-engineer/SKILL.md +12 -4
  73. package/.agent/skills/devops-incident-responder/SKILL.md +12 -4
  74. package/.agent/skills/doc.md +1 -1
  75. package/.agent/skills/documentation-templates/SKILL.md +12 -4
  76. package/.agent/skills/edge-computing/SKILL.md +12 -4
  77. package/.agent/skills/error-resilience/SKILL.md +390 -382
  78. package/.agent/skills/extract-design-system/SKILL.md +12 -4
  79. package/.agent/skills/framer-motion-expert/SKILL.md +206 -199
  80. package/.agent/skills/frontend-design/SKILL.md +163 -155
  81. package/.agent/skills/game-design-expert/SKILL.md +12 -4
  82. package/.agent/skills/game-engineering-expert/SKILL.md +12 -4
  83. package/.agent/skills/geo-fundamentals/SKILL.md +12 -4
  84. package/.agent/skills/github-operations/SKILL.md +12 -4
  85. package/.agent/skills/gsap-core/SKILL.md +54 -48
  86. package/.agent/skills/gsap-frameworks/SKILL.md +54 -48
  87. package/.agent/skills/gsap-performance/SKILL.md +54 -48
  88. package/.agent/skills/gsap-plugins/SKILL.md +54 -48
  89. package/.agent/skills/gsap-react/SKILL.md +54 -48
  90. package/.agent/skills/gsap-scrolltrigger/SKILL.md +54 -48
  91. package/.agent/skills/gsap-timeline/SKILL.md +54 -48
  92. package/.agent/skills/gsap-utils/SKILL.md +54 -48
  93. package/.agent/skills/i18n-localization/SKILL.md +12 -4
  94. package/.agent/skills/intelligent-routing/SKILL.md +41 -33
  95. package/.agent/skills/knowledge-graph/SKILL.md +36 -0
  96. package/.agent/skills/lint-and-validate/SKILL.md +12 -4
  97. package/.agent/skills/llm-engineering/SKILL.md +12 -4
  98. package/.agent/skills/local-first/SKILL.md +12 -4
  99. package/.agent/skills/mcp-builder/SKILL.md +12 -4
  100. package/.agent/skills/mobile-design/SKILL.md +225 -217
  101. package/.agent/skills/monorepo-management/SKILL.md +296 -288
  102. package/.agent/skills/motion-engineering/SKILL.md +195 -187
  103. package/.agent/skills/nextjs-react-expert/SKILL.md +196 -188
  104. package/.agent/skills/nodejs-best-practices/SKILL.md +12 -4
  105. package/.agent/skills/observability/SKILL.md +12 -4
  106. package/.agent/skills/parallel-agents/SKILL.md +12 -4
  107. package/.agent/skills/performance-profiling/SKILL.md +12 -4
  108. package/.agent/skills/plan-writing/SKILL.md +12 -4
  109. package/.agent/skills/platform-engineer/SKILL.md +12 -4
  110. package/.agent/skills/playwright-best-practices/SKILL.md +12 -4
  111. package/.agent/skills/powershell-windows/SKILL.md +12 -4
  112. package/.agent/skills/project-idioms/SKILL.md +12 -4
  113. package/.agent/skills/python-patterns/SKILL.md +12 -4
  114. package/.agent/skills/python-pro/SKILL.md +285 -277
  115. package/.agent/skills/react-specialist/SKILL.md +239 -231
  116. package/.agent/skills/readme-builder/SKILL.md +12 -4
  117. package/.agent/skills/realtime-patterns/SKILL.md +12 -4
  118. package/.agent/skills/red-team-tactics/SKILL.md +12 -4
  119. package/.agent/skills/rust-pro/SKILL.md +12 -4
  120. package/.agent/skills/seo-fundamentals/SKILL.md +12 -4
  121. package/.agent/skills/server-management/SKILL.md +12 -4
  122. package/.agent/skills/shadcn-ui-expert/SKILL.md +12 -4
  123. package/.agent/skills/skill-creator/SKILL.md +12 -4
  124. package/.agent/skills/sql-pro/SKILL.md +12 -4
  125. package/.agent/skills/supabase-postgres-best-practices/SKILL.md +12 -4
  126. package/.agent/skills/swiftui-expert/SKILL.md +12 -4
  127. package/.agent/skills/systematic-debugging/SKILL.md +12 -4
  128. package/.agent/skills/tailwind-patterns/SKILL.md +12 -4
  129. package/.agent/skills/tdd-workflow/SKILL.md +12 -4
  130. package/.agent/skills/test-result-analyzer/SKILL.md +12 -4
  131. package/.agent/skills/testing-patterns/SKILL.md +12 -4
  132. package/.agent/skills/trend-researcher/SKILL.md +12 -4
  133. package/.agent/skills/typescript-advanced/SKILL.md +297 -289
  134. package/.agent/skills/ui-ux-pro-max/SKILL.md +12 -4
  135. package/.agent/skills/ui-ux-researcher/SKILL.md +12 -4
  136. package/.agent/skills/vue-expert/SKILL.md +237 -229
  137. package/.agent/skills/vulnerability-scanner/SKILL.md +12 -4
  138. package/.agent/skills/web-accessibility-auditor/SKILL.md +12 -4
  139. package/.agent/skills/web-design-guidelines/SKILL.md +12 -4
  140. package/.agent/skills/webapp-testing/SKILL.md +12 -4
  141. package/.agent/skills/whimsy-injector/SKILL.md +12 -4
  142. package/.agent/skills/workflow-optimizer/SKILL.md +12 -4
  143. package/.agent/workflows/audit.md +6 -6
  144. package/.agent/workflows/deploy.md +1 -1
  145. package/.agent/workflows/generate.md +23 -6
  146. package/.agent/workflows/session.md +5 -5
  147. package/.agent/workflows/swarm.md +2 -2
  148. package/README.md +242 -186
  149. package/bin/tribunal-kit.js +297 -57
  150. package/package.json +81 -77
  151. package/scripts/changelog.js +167 -0
  152. package/scripts/sync-version.js +81 -0
  153. package/scripts/validate-payload.js +73 -0
  154. package/.agent/scripts/__pycache__/auto_preview.cpython-311.pyc +0 -0
  155. package/.agent/scripts/__pycache__/bundle_analyzer.cpython-311.pyc +0 -0
  156. package/.agent/scripts/__pycache__/checklist.cpython-311.pyc +0 -0
  157. package/.agent/scripts/__pycache__/dependency_analyzer.cpython-311.pyc +0 -0
  158. package/.agent/scripts/__pycache__/security_scan.cpython-311.pyc +0 -0
  159. package/.agent/scripts/__pycache__/session_manager.cpython-311.pyc +0 -0
  160. package/.agent/scripts/__pycache__/skill_integrator.cpython-311.pyc +0 -0
  161. package/.agent/scripts/__pycache__/swarm_dispatcher.cpython-311.pyc +0 -0
  162. package/.agent/scripts/__pycache__/test_runner.cpython-311.pyc +0 -0
  163. package/.agent/scripts/__pycache__/verify_all.cpython-311.pyc +0 -0
  164. package/.agent/scripts/auto_preview.py +0 -180
  165. package/.agent/scripts/bundle_analyzer.py +0 -259
  166. package/.agent/scripts/case_law_manager.py +0 -755
  167. package/.agent/scripts/checklist.py +0 -209
  168. package/.agent/scripts/compress_skills.py +0 -167
  169. package/.agent/scripts/consolidate_skills.py +0 -173
  170. package/.agent/scripts/deep_compress.py +0 -202
  171. package/.agent/scripts/dependency_analyzer.py +0 -247
  172. package/.agent/scripts/lint_runner.py +0 -188
  173. package/.agent/scripts/minify_context.py +0 -80
  174. package/.agent/scripts/patch_skills_meta.py +0 -177
  175. package/.agent/scripts/patch_skills_output.py +0 -285
  176. package/.agent/scripts/schema_validator.py +0 -279
  177. package/.agent/scripts/security_scan.py +0 -224
  178. package/.agent/scripts/session_manager.py +0 -261
  179. package/.agent/scripts/skill_evolution.py +0 -563
  180. package/.agent/scripts/skill_integrator.py +0 -234
  181. package/.agent/scripts/strengthen_skills.py +0 -220
  182. package/.agent/scripts/strip_tribunal.py +0 -41
  183. package/.agent/scripts/swarm_dispatcher.py +0 -350
  184. package/.agent/scripts/test_runner.py +0 -192
  185. package/.agent/scripts/test_swarm_dispatcher.py +0 -163
  186. package/.agent/scripts/verify_all.py +0 -195
@@ -1,286 +1,286 @@
1
- ---
2
- name: data-validation-schemas
3
- description: Data validation and schema design mastery. Zod, Yup, Joi, Valibot, and Pydantic schema design, runtime type checking, API boundary validation, form validation patterns, DTO design, schema composition, error message formatting, schema evolution strategies, and coercion rules. Use when validating user input, API payloads, environment config, or any data crossing a trust boundary.
4
- allowed-tools: Read, Write, Edit, Glob, Grep
5
- version: 1.0.0
6
- last-updated: 2026-04-17
7
- applies-to-model: gemini-2.5-pro, claude-3-7-sonnet
8
- ---
9
-
10
- ## Hallucination Traps (Read First)
11
- - ❌ Using `z.any()` or `z.unknown()` as a lazy escape hatch -> ✅ Always define the actual shape; `any` defeats the purpose of validation
12
- - ❌ Validating on the client but not on the server -> ✅ Server validation is NOT optional — client validation is UX, server validation is security
13
- - ❌ Throwing raw Zod errors to the client -> ✅ Format errors into user-friendly messages with `.flatten()` or `.format()`
14
-
15
- ---
16
-
17
- # Data Validation & Schemas — Trust No Input
18
-
19
- ---
20
-
21
- ## The Golden Rule
22
-
23
- ```
24
- Every trust boundary gets a schema.
25
- No exceptions. No shortcuts. No "I'll add validation later."
26
-
27
- Trust Boundaries:
28
- ✅ API request bodies (user → server)
29
- ✅ URL params / query strings (user → server)
30
- ✅ Environment variables (env → app)
31
- ✅ External API responses (3rd party → app)
32
- ✅ Database query results (DB → app, if untyped)
33
- ✅ File uploads (user → server)
34
- ✅ WebSocket messages (client → server)
35
- ✅ Form inputs (user → UI)
36
- ```
37
-
38
- ---
39
-
40
- ## Zod (Recommended — TypeScript)
41
-
42
- ### Basic Schemas
43
-
44
- ```typescript
45
- import { z } from "zod";
46
-
47
- // Primitives with constraints
48
- const Email = z.string().email().toLowerCase().trim();
49
- const Age = z.number().int().min(0).max(150);
50
- const Username = z.string().min(3).max(30).regex(/^[a-zA-Z0-9_]+$/);
51
- const URL = z.string().url().startsWith("https://");
52
-
53
- // Object schema
54
- const CreateUserSchema = z.object({
55
- name: z.string().min(2).max(100),
56
- email: Email,
57
- age: Age.optional(),
58
- role: z.enum(["admin", "editor", "viewer"]).default("viewer"),
59
- metadata: z.record(z.string(), z.unknown()).optional(),
60
- });
61
-
62
- // ✅ Infer TypeScript types from schemas (single source of truth)
63
- type CreateUserInput = z.infer<typeof CreateUserSchema>;
64
- // → { name: string; email: string; age?: number; role: "admin" | "editor" | "viewer"; ... }
65
- ```
66
-
67
- ### Composition & Reuse
68
-
69
- ```typescript
70
- // ✅ Base schema + extend for variants
71
- const BaseUserSchema = z.object({
72
- name: z.string().min(2),
73
- email: z.string().email(),
74
- });
75
-
76
- const CreateUserSchema = BaseUserSchema.extend({
77
- password: z.string().min(8),
78
- confirmPassword: z.string(),
79
- }).refine((data) => data.password === data.confirmPassword, {
80
- message: "Passwords don't match",
81
- path: ["confirmPassword"],
82
- });
83
-
84
- const UpdateUserSchema = BaseUserSchema.partial(); // all fields optional
85
-
86
- // ✅ Pick / Omit
87
- const LoginSchema = BaseUserSchema.pick({ email: true }).extend({
88
- password: z.string(),
89
- });
90
-
91
- // ✅ Merge two schemas
92
- const FullProfileSchema = BaseUserSchema.merge(AddressSchema);
93
- ```
94
-
95
- ### API Boundary Validation
96
-
97
- ```typescript
98
- // ✅ Server-side: validate at the boundary, type-safe downstream
99
- import { z } from "zod";
100
-
101
- // Define once, use everywhere
102
- const QuerySchema = z.object({
103
- page: z.coerce.number().int().min(1).default(1),
104
- limit: z.coerce.number().int().min(1).max(100).default(20),
105
- sort: z.enum(["created", "updated", "name"]).default("created"),
106
- order: z.enum(["asc", "desc"]).default("desc"),
107
- search: z.string().max(200).optional(),
108
- });
109
-
110
- // Express middleware
111
- function validate<T extends z.ZodType>(schema: T) {
112
- return (req: Request, res: Response, next: NextFunction) => {
113
- const result = schema.safeParse(req.body);
114
- if (!result.success) {
115
- return res.status(400).json({
116
- error: "Validation failed",
117
- issues: result.error.flatten().fieldErrors,
118
- });
119
- }
120
- req.body = result.data; // ✅ Validated + coerced data replaces raw body
121
- next();
122
- };
123
- }
124
-
125
- app.post("/api/users", validate(CreateUserSchema), async (req, res) => {
126
- // req.body is now fully typed and validated
127
- const user = await createUser(req.body);
128
- res.status(201).json(user);
129
- });
130
-
131
- // ❌ BAD: Validating inside the handler
132
- // ✅ GOOD: Validation as middleware — keeps handlers clean
133
- ```
134
-
135
- ### Error Formatting
136
-
137
- ```typescript
138
- // ✅ User-friendly error messages
139
- const result = CreateUserSchema.safeParse(rawInput);
140
-
141
- if (!result.success) {
142
- // .flatten() — flat structure for simple forms
143
- const flat = result.error.flatten();
144
- // { fieldErrors: { email: ["Invalid email"], name: ["Too short"] } }
145
-
146
- // .format() — nested structure matching schema shape
147
- const formatted = result.error.format();
148
- // { email: { _errors: ["Invalid email"] }, name: { _errors: ["Too short"] } }
149
-
150
- // Custom error map (global)
151
- z.setErrorMap((issue, ctx) => {
152
- if (issue.code === z.ZodIssueCode.too_small) {
153
- return { message: `Must be at least ${issue.minimum} characters` };
154
- }
155
- return { message: ctx.defaultError };
156
- });
157
- }
158
- ```
159
-
160
- ---
161
-
162
- ## Environment Validation (Fail Fast)
163
-
164
- ```typescript
165
- // ✅ Validate ALL env vars at startup — crash immediately if invalid
166
- const EnvSchema = z.object({
167
- NODE_ENV: z.enum(["development", "production", "test"]),
168
- PORT: z.coerce.number().default(3000),
169
- DATABASE_URL: z.string().url(),
170
- REDIS_URL: z.string().url().optional(),
171
- JWT_SECRET: z.string().min(32, "JWT_SECRET must be ≥ 32 characters"),
172
- API_KEY: z.string().min(1),
173
- LOG_LEVEL: z.enum(["debug", "info", "warn", "error"]).default("info"),
174
- });
175
-
176
- export const env = EnvSchema.parse(process.env);
177
-
178
- // ❌ TRAP: process.env.DATABASE_URL! ← crashes at RUNTIME, not startup
179
- // ✅ Parse at module load → crash at STARTUP with clear error message
180
- ```
181
-
182
- ---
183
-
184
- ## Pydantic (Python)
185
-
186
- ```python
187
- from pydantic import BaseModel, Field, field_validator, model_validator
188
- from datetime import datetime
189
-
190
- class CreateUserRequest(BaseModel):
191
- name: str = Field(min_length=2, max_length=100)
192
- email: str = Field(pattern=r"^[\w\.\+\-]+@[\w]+\.[\w\.]+$")
193
- age: int | None = Field(default=None, ge=0, le=150)
194
- role: str = Field(default="viewer")
195
-
196
- @field_validator("email")
197
- @classmethod
198
- def normalize_email(cls, v: str) -> str:
199
- return v.lower().strip()
200
-
201
- @field_validator("role")
202
- @classmethod
203
- def validate_role(cls, v: str) -> str:
204
- allowed = {"admin", "editor", "viewer"}
205
- if v not in allowed:
206
- raise ValueError(f"Role must be one of: {allowed}")
207
- return v
208
-
209
- # FastAPI uses Pydantic automatically
210
- @app.post("/users")
211
- async def create_user(user: CreateUserRequest):
212
- # user is already validated and typed
213
- return await db.create_user(user.model_dump())
214
- ```
215
-
216
- ---
217
-
218
- ## Form Validation (React + Zod)
219
-
220
- ```tsx
221
- // ✅ React Hook Form + Zod = type-safe forms
222
- import { useForm } from "react-hook-form";
223
- import { zodResolver } from "@hookform/resolvers/zod";
224
- import { z } from "zod";
225
-
226
- const SignupSchema = z.object({
227
- email: z.string().email("Invalid email address"),
228
- password: z.string()
229
- .min(8, "Password must be at least 8 characters")
230
- .regex(/[A-Z]/, "Must contain uppercase letter")
231
- .regex(/[0-9]/, "Must contain a number"),
232
- terms: z.literal(true, {
233
- errorMap: () => ({ message: "You must accept the terms" }),
234
- }),
235
- });
236
-
237
- type SignupData = z.infer<typeof SignupSchema>;
238
-
239
- function SignupForm() {
240
- const { register, handleSubmit, formState: { errors } } = useForm<SignupData>({
241
- resolver: zodResolver(SignupSchema),
242
- });
243
-
244
- return (
245
- <form onSubmit={handleSubmit((data) => signup(data))}>
246
- <input {...register("email")} />
247
- {errors.email && <span>{errors.email.message}</span>}
248
-
249
- <input type="password" {...register("password")} />
250
- {errors.password && <span>{errors.password.message}</span>}
251
-
252
- <label>
253
- <input type="checkbox" {...register("terms")} />
254
- I accept the terms
255
- </label>
256
- {errors.terms && <span>{errors.terms.message}</span>}
257
-
258
- <button type="submit">Sign Up</button>
259
- </form>
260
- );
261
- }
262
- ```
263
-
264
- ---
265
-
266
- ## Schema Anti-Patterns
267
-
268
- ```
269
- ❌ z.any() / z.unknown() as a lazy escape — defeats the purpose
270
- ❌ Validating on client only — server is the security boundary
271
- ❌ Different schemas for same entity on client vs server — drift guaranteed
272
- ❌ Coercing without documenting — z.coerce.number() silently converts "abc" → NaN
273
- ❌ Skipping .safeParse() in user-facing code — .parse() throws, bad UX
274
- ❌ Giant monolithic schemas — use .extend(), .pick(), .merge() for composition
275
- ❌ Not validating 3rd-party API responses — "they'll always return what docs say"
276
- ```
277
-
278
- ---
1
+ ---
2
+ name: data-validation-schemas
3
+ description: Data validation and schema design mastery. Zod, Yup, Joi, Valibot, and Pydantic schema design, runtime type checking, API boundary validation, form validation patterns, DTO design, schema composition, error message formatting, schema evolution strategies, and coercion rules. Use when validating user input, API payloads, environment config, or any data crossing a trust boundary.
4
+ allowed-tools: Read, Write, Edit, Glob, Grep
5
+ version: 1.0.0
6
+ last-updated: 2026-04-17
7
+ applies-to-model: gemini-2.5-pro, claude-3-7-sonnet
8
+ ---
9
+
10
+ ## Hallucination Traps (Read First)
11
+ - ❌ Using `z.any()` or `z.unknown()` as a lazy escape hatch -> ✅ Always define the actual shape; `any` defeats the purpose of validation
12
+ - ❌ Validating on the client but not on the server -> ✅ Server validation is NOT optional — client validation is UX, server validation is security
13
+ - ❌ Throwing raw Zod errors to the client -> ✅ Format errors into user-friendly messages with `.flatten()` or `.format()`
14
+
15
+ ---
16
+
17
+ # Data Validation & Schemas — Trust No Input
18
+
19
+ ---
20
+
21
+ ## The Golden Rule
22
+
23
+ ```
24
+ Every trust boundary gets a schema.
25
+ No exceptions. No shortcuts. No "I'll add validation later."
26
+
27
+ Trust Boundaries:
28
+ ✅ API request bodies (user → server)
29
+ ✅ URL params / query strings (user → server)
30
+ ✅ Environment variables (env → app)
31
+ ✅ External API responses (3rd party → app)
32
+ ✅ Database query results (DB → app, if untyped)
33
+ ✅ File uploads (user → server)
34
+ ✅ WebSocket messages (client → server)
35
+ ✅ Form inputs (user → UI)
36
+ ```
37
+
38
+ ---
39
+
40
+ ## Zod (Recommended — TypeScript)
41
+
42
+ ### Basic Schemas
43
+
44
+ ```typescript
45
+ import { z } from "zod";
46
+
47
+ // Primitives with constraints
48
+ const Email = z.string().email().toLowerCase().trim();
49
+ const Age = z.number().int().min(0).max(150);
50
+ const Username = z.string().min(3).max(30).regex(/^[a-zA-Z0-9_]+$/);
51
+ const URL = z.string().url().startsWith("https://");
52
+
53
+ // Object schema
54
+ const CreateUserSchema = z.object({
55
+ name: z.string().min(2).max(100),
56
+ email: Email,
57
+ age: Age.optional(),
58
+ role: z.enum(["admin", "editor", "viewer"]).default("viewer"),
59
+ metadata: z.record(z.string(), z.unknown()).optional(),
60
+ });
61
+
62
+ // ✅ Infer TypeScript types from schemas (single source of truth)
63
+ type CreateUserInput = z.infer<typeof CreateUserSchema>;
64
+ // → { name: string; email: string; age?: number; role: "admin" | "editor" | "viewer"; ... }
65
+ ```
66
+
67
+ ### Composition & Reuse
68
+
69
+ ```typescript
70
+ // ✅ Base schema + extend for variants
71
+ const BaseUserSchema = z.object({
72
+ name: z.string().min(2),
73
+ email: z.string().email(),
74
+ });
75
+
76
+ const CreateUserSchema = BaseUserSchema.extend({
77
+ password: z.string().min(8),
78
+ confirmPassword: z.string(),
79
+ }).refine((data) => data.password === data.confirmPassword, {
80
+ message: "Passwords don't match",
81
+ path: ["confirmPassword"],
82
+ });
83
+
84
+ const UpdateUserSchema = BaseUserSchema.partial(); // all fields optional
85
+
86
+ // ✅ Pick / Omit
87
+ const LoginSchema = BaseUserSchema.pick({ email: true }).extend({
88
+ password: z.string(),
89
+ });
90
+
91
+ // ✅ Merge two schemas
92
+ const FullProfileSchema = BaseUserSchema.merge(AddressSchema);
93
+ ```
94
+
95
+ ### API Boundary Validation
96
+
97
+ ```typescript
98
+ // ✅ Server-side: validate at the boundary, type-safe downstream
99
+ import { z } from "zod";
100
+
101
+ // Define once, use everywhere
102
+ const QuerySchema = z.object({
103
+ page: z.coerce.number().int().min(1).default(1),
104
+ limit: z.coerce.number().int().min(1).max(100).default(20),
105
+ sort: z.enum(["created", "updated", "name"]).default("created"),
106
+ order: z.enum(["asc", "desc"]).default("desc"),
107
+ search: z.string().max(200).optional(),
108
+ });
109
+
110
+ // Express middleware
111
+ function validate<T extends z.ZodType>(schema: T) {
112
+ return (req: Request, res: Response, next: NextFunction) => {
113
+ const result = schema.safeParse(req.body);
114
+ if (!result.success) {
115
+ return res.status(400).json({
116
+ error: "Validation failed",
117
+ issues: result.error.flatten().fieldErrors,
118
+ });
119
+ }
120
+ req.body = result.data; // ✅ Validated + coerced data replaces raw body
121
+ next();
122
+ };
123
+ }
124
+
125
+ app.post("/api/users", validate(CreateUserSchema), async (req, res) => {
126
+ // req.body is now fully typed and validated
127
+ const user = await createUser(req.body);
128
+ res.status(201).json(user);
129
+ });
130
+
131
+ // ❌ BAD: Validating inside the handler
132
+ // ✅ GOOD: Validation as middleware — keeps handlers clean
133
+ ```
134
+
135
+ ### Error Formatting
136
+
137
+ ```typescript
138
+ // ✅ User-friendly error messages
139
+ const result = CreateUserSchema.safeParse(rawInput);
140
+
141
+ if (!result.success) {
142
+ // .flatten() — flat structure for simple forms
143
+ const flat = result.error.flatten();
144
+ // { fieldErrors: { email: ["Invalid email"], name: ["Too short"] } }
145
+
146
+ // .format() — nested structure matching schema shape
147
+ const formatted = result.error.format();
148
+ // { email: { _errors: ["Invalid email"] }, name: { _errors: ["Too short"] } }
149
+
150
+ // Custom error map (global)
151
+ z.setErrorMap((issue, ctx) => {
152
+ if (issue.code === z.ZodIssueCode.too_small) {
153
+ return { message: `Must be at least ${issue.minimum} characters` };
154
+ }
155
+ return { message: ctx.defaultError };
156
+ });
157
+ }
158
+ ```
279
159
 
160
+ ---
161
+
162
+ ## Environment Validation (Fail Fast)
163
+
164
+ ```typescript
165
+ // ✅ Validate ALL env vars at startup — crash immediately if invalid
166
+ const EnvSchema = z.object({
167
+ NODE_ENV: z.enum(["development", "production", "test"]),
168
+ PORT: z.coerce.number().default(3000),
169
+ DATABASE_URL: z.string().url(),
170
+ REDIS_URL: z.string().url().optional(),
171
+ JWT_SECRET: z.string().min(32, "JWT_SECRET must be ≥ 32 characters"),
172
+ API_KEY: z.string().min(1),
173
+ LOG_LEVEL: z.enum(["debug", "info", "warn", "error"]).default("info"),
174
+ });
175
+
176
+ export const env = EnvSchema.parse(process.env);
177
+
178
+ // ❌ TRAP: process.env.DATABASE_URL! ← crashes at RUNTIME, not startup
179
+ // ✅ Parse at module load → crash at STARTUP with clear error message
180
+ ```
181
+
182
+ ---
183
+
184
+ ## Pydantic (Python)
185
+
186
+ ```python
187
+ from pydantic import BaseModel, Field, field_validator, model_validator
188
+ from datetime import datetime
189
+
190
+ class CreateUserRequest(BaseModel):
191
+ name: str = Field(min_length=2, max_length=100)
192
+ email: str = Field(pattern=r"^[\w\.\+\-]+@[\w]+\.[\w\.]+$")
193
+ age: int | None = Field(default=None, ge=0, le=150)
194
+ role: str = Field(default="viewer")
195
+
196
+ @field_validator("email")
197
+ @classmethod
198
+ def normalize_email(cls, v: str) -> str:
199
+ return v.lower().strip()
200
+
201
+ @field_validator("role")
202
+ @classmethod
203
+ def validate_role(cls, v: str) -> str:
204
+ allowed = {"admin", "editor", "viewer"}
205
+ if v not in allowed:
206
+ raise ValueError(f"Role must be one of: {allowed}")
207
+ return v
208
+
209
+ # FastAPI uses Pydantic automatically
210
+ @app.post("/users")
211
+ async def create_user(user: CreateUserRequest):
212
+ # user is already validated and typed
213
+ return await db.create_user(user.model_dump())
214
+ ```
215
+
216
+ ---
217
+
218
+ ## Form Validation (React + Zod)
219
+
220
+ ```tsx
221
+ // ✅ React Hook Form + Zod = type-safe forms
222
+ import { useForm } from "react-hook-form";
223
+ import { zodResolver } from "@hookform/resolvers/zod";
224
+ import { z } from "zod";
225
+
226
+ const SignupSchema = z.object({
227
+ email: z.string().email("Invalid email address"),
228
+ password: z.string()
229
+ .min(8, "Password must be at least 8 characters")
230
+ .regex(/[A-Z]/, "Must contain uppercase letter")
231
+ .regex(/[0-9]/, "Must contain a number"),
232
+ terms: z.literal(true, {
233
+ errorMap: () => ({ message: "You must accept the terms" }),
234
+ }),
235
+ });
236
+
237
+ type SignupData = z.infer<typeof SignupSchema>;
238
+
239
+ function SignupForm() {
240
+ const { register, handleSubmit, formState: { errors } } = useForm<SignupData>({
241
+ resolver: zodResolver(SignupSchema),
242
+ });
243
+
244
+ return (
245
+ <form onSubmit={handleSubmit((data) => signup(data))}>
246
+ <input {...register("email")} />
247
+ {errors.email && <span>{errors.email.message}</span>}
248
+
249
+ <input type="password" {...register("password")} />
250
+ {errors.password && <span>{errors.password.message}</span>}
251
+
252
+ <label>
253
+ <input type="checkbox" {...register("terms")} />
254
+ I accept the terms
255
+ </label>
256
+ {errors.terms && <span>{errors.terms.message}</span>}
257
+
258
+ <button type="submit">Sign Up</button>
259
+ </form>
260
+ );
261
+ }
262
+ ```
280
263
 
281
264
  ---
282
265
 
283
- ## 🤖 LLM-Specific Traps
266
+ ## Schema Anti-Patterns
267
+
268
+ ```
269
+ ❌ z.any() / z.unknown() as a lazy escape — defeats the purpose
270
+ ❌ Validating on client only — server is the security boundary
271
+ ❌ Different schemas for same entity on client vs server — drift guaranteed
272
+ ❌ Coercing without documenting — z.coerce.number() silently converts "abc" → NaN
273
+ ❌ Skipping .safeParse() in user-facing code — .parse() throws, bad UX
274
+ ❌ Giant monolithic schemas — use .extend(), .pick(), .merge() for composition
275
+ ❌ Not validating 3rd-party API responses — "they'll always return what docs say"
276
+ ```
277
+
278
+ ---
279
+
280
+
281
+ ---
282
+
283
+
284
284
 
285
285
  AI coding assistants often fall into specific bad habits when dealing with this domain. These are strictly forbidden:
286
286
 
@@ -292,7 +292,7 @@ AI coding assistants often fall into specific bad habits when dealing with this
292
292
 
293
293
  ---
294
294
 
295
- ## 🏛️ Tribunal Integration (Anti-Hallucination)
295
+
296
296
 
297
297
  **Slash command: `/review` or `/tribunal-full`**
298
298
  **Active reviewers: `logic-reviewer` · `security-auditor`**
@@ -303,7 +303,7 @@ AI coding assistants often fall into specific bad habits when dealing with this
303
303
  2. **Silent Degradation:** Catching and suppressing errors without logging or handling.
304
304
  3. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
305
305
 
306
- ### ✅ Pre-Flight Self-Audit
306
+
307
307
 
308
308
  Review these questions before confirming output:
309
309
  ```
@@ -317,4 +317,12 @@ Review these questions before confirming output:
317
317
 
318
318
  **CRITICAL:** You must follow a strict "evidence-based closeout" state machine.
319
319
  - ❌ **Forbidden:** Declaring a task complete because the output "looks correct."
320
- - ✅ **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.
320
+ - ✅ **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.
321
+
322
+
323
+ ## Pre-Flight Checklist
324
+ - [ ] Have I reviewed the user's specific constraints and requests?
325
+ - [ ] Have I checked the environment for relevant existing implementations?
326
+
327
+ ## VBC Protocol (Verification-Before-Completion)
328
+ You MUST verify existing code signatures and variables before attempting to modify or call them. No hallucination is permitted.