ari-zod-schemas
Zod schema management for ARI's type-safe runtime validation
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Menu
Zod schema management for ARI's type-safe runtime validation
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Discord slash commands, approval routing, channel policy, button interaction patterns for OpenClaw/ARI Discord integration
Obsidian vault integration patterns — vault-analyzer.ts, /ari-vault-* commands, morning briefing snippet, PARA structure, read-only enforcement
OpenClaw plugin development patterns — hooks, manifest structure, plugin SDK, APEX/CODEX enforcement
NOVA's P1 PayThePryce pipeline — market signal ingest, card detection, price monitoring, script generation, thumbnail generation, video assembly, approval gate
CHASE's P2 Pryceless Solutions pipeline — lead discovery, 5-criteria audit, LLM qualification, Prompt Forge 4-pass lock, demo generation, outreach approval gate
NOVA's thumbnail generation pipeline — Ideogram V3 via Fal.ai (primary) + DALL-E 3 fallback, 4-variant strategy, Pokemon TCG copyright rules,
| name | ari-zod-schemas |
| description | Zod schema management for ARI's type-safe runtime validation |
| triggers | ["create schema","validate types","zod schema","add config validation"] |
ARI uses Zod for all config and data structure validation (ADR-006). This skill ensures:
All Zod schemas live in: src/kernel/types.ts
// Trust levels with risk multipliers
const TrustLevelSchema = z.enum([
'SYSTEM', // 0.5x
'OPERATOR', // 0.6x
'VERIFIED', // 0.75x
'STANDARD', // 1.0x
'UNTRUSTED', // 1.5x
'HOSTILE' // 2.0x
]);
// Audit event for hash chain
const AuditEventSchema = z.object({
id: z.string().uuid(),
timestamp: z.string().datetime(),
action: z.string(),
agent: z.string().optional(),
details: z.record(z.unknown()),
previousHash: z.string(),
hash: z.string()
});
// Message schema
const MessageSchema = z.object({
id: z.string().uuid(),
content: z.string(),
trustLevel: TrustLevelSchema,
timestamp: z.string().datetime(),
metadata: z.record(z.unknown()).optional()
});
// 1. Define schema
export const NewFeatureSchema = z.object({
name: z.string().min(1),
enabled: z.boolean().default(true),
config: z.record(z.string())
});
// 2. Export type
export type NewFeature = z.infer<typeof NewFeatureSchema>;
// 3. Validate at boundary
const validated = NewFeatureSchema.parse(untrustedInput);
const EnvSchema = z.object({
ARI_PORT: z.coerce.number().default(3141),
ARI_LOG_LEVEL: z.enum(['debug', 'info', 'warn', 'error']).default('info'),
// Add new vars here FIRST
});
const EventPayloadSchema = z.discriminatedUnion('type', [
z.object({ type: z.literal('message'), data: MessageSchema }),
z.object({ type: z.literal('audit'), data: AuditEventSchema }),
]);
const ConfigSchema = z.object({
gateway: z.object({
host: z.literal('127.0.0.1'), // Enforces loopback-only
port: z.number().min(1024).max(65535)
}),
security: z.object({
maxRiskScore: z.number().min(0).max(1).default(0.8)
})
});
.strict() to reject unknown keys