Skip to main content

Explain evaluation

GET 

/api/policy-evaluations/:evaluation_id/explain

The explanation is derived from the RECORDED decision, never re-guessed by an LLM (ยง21). When the evaluation predates context_snapshot (POL migration 19), the context is reconstructed from the entity's CURRENT state instead โ€” reconstructed: true on the response says so explicitly rather than silently presenting a best-effort trace as an exact historical one.

Requestโ€‹

Responsesโ€‹

Successful Response