Propose a prompt fix for an eval regression
Given a measured regression — an eval-suite pass-rate, quality scorecard, or experiment metric that dropped versus the prior version, plus a sample of the failing cases — generate a root-cause analysis and a revised system prompt, and mint it as a candidate branch version. Apply it through the existing promotion path or a A/B experiment; the endpoints are echoed back in suggested_actions. Owner, admin, and developer roles only (it burns LLM tokens and mints a version).
Authorizations
Dashboard JWT token from Clerk
Headers
Stripe-style idempotency token. Pass a stable, client-generated value (1-255 chars) to dedupe retries on transient timeouts. The same key+credential+path replays the original response for 24h on 2xx (5min on 4xx, 30s on 5xx). Returns 409 if a concurrent request with the same key is already in flight; replayed responses include the Idempotency-Replay: true response header.
1 - 255Sandbox opt-in for Clerk-session-authenticated requests. Set to true to route the call through the test-mode pipeline: no real provider delivery, no credits deducted, response meta.test_mode: true. Ignored for live API keys (dv_live_sk_*) — server-to-server clients must use a test-prefixed key (dv_test_sk_*) to exercise sandbox. Test-prefixed keys unconditionally enable sandbox regardless of this header.
true, false Path Parameters
Response
Successful response.