AITracer
AITracer brings the original dark Audit Vault identity into a cleaner architecture. AITracer exists for one reason: production AI systems need proof, not guesswork. Ordinary logs can show that a request happened. AITracer goes further by connecting model metadata, token usage, cost, latency, policy results, and SHA-256 verification data into
AI
Production rollout should include: * verified DATABASE_URL and migration readiness * Auth0 callback domain and audience validation * Stripe checkout/portal keys scoped to the target environment * ingest API smoke tests for trace, policy, verification, and audit creation * dashboard checks for alerts, cost, and governance visibility These checks reduce launch risk and
AI
The Audit Vault is AITracer's governance center. It gives compliance, product, engineering, and security teams a shared record of what happened during an AI execution. What belongs in the vault A useful AI execution record should capture: * request and response metadata, * model and workflow identifiers, * input and output
AI
Preview deployments are useful for validating policy controls before production rollout. Recommended preview checks * ingest a controlled trace payload through /api/traces * verify Trace, ExecutionRecord, and VerificationRecord inserts * confirm governance outcomes for known sensitive patterns * verify audit export visibility and retention controls Parity requirements Preview environments should mirror production policy