Read one conversation's execution trace
Return the audit-grade execution trace for one conversation: every LLM turn’s prompts, assistant response, token usage (prompt / completion / cache), latency, model and version, plus the tool dispatches and retrievals, with tool calls that were not attributable to a returned turn surfaced separately. audit_enabled: false means the tenant’s governance log is disabled, and the trace degrades to unmatched tool activity only. Elevated roles only, because verbatim prompts and tool payloads may carry redacted PII.
Authorizations
Dashboard JWT token from Clerk
Headers
Sandbox 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
The agent id.
1The agent conversation id.
1 - 200