Which metadata is stored, and how content logging and retention work.
By default Parezen stores metadata only: organisation, project, environment, key id, route, task, agent, workflow, trace, model, token counts, cost, latency, status and which levers ran. That is enough for the ledger.
Prompt and response content stays off. Content logging is a per-project setting on Inference and stays off unless you turn it on; sensitive values are redacted before storage, and the change is written to the audit log. Observability metadata and OpenTelemetry (OTLP) export are part of Inference. Pages that show a prompt show a locked state instead when logging is off.
Request logs are kept about 30 days on the hosted gateway. Prompt and reply content is kept only for workspaces that turn content logging on, for the retention period you set, and is then deleted. Human review of a replay sample happens only after a written opt-in. Do not treat this page as a data-processing agreement.
Questions go to [email protected]. The public policy is the privacy policy.
