Skip to main content
POST
Evaluate a card against an agent's historical traces

Authorizations

Authorization
string
header
required

Supabase JWT token in Authorization: Bearer header

Body

application/json
agent_id
string
required

Agent whose historical traces will be replayed. The caller must be an owner, admin or member of the agent's org; otherwise the response is the same 404 as for an unknown agent.

card_json
object

Optional hypothetical alignment card to evaluate against. If omitted, the agent's current canonical card is used (404 if the agent has none and no card_json is supplied).

limit
integer
default:50

Maximum number of recent traces to evaluate (capped at 200).

Required range: x <= 200

Response

Historical evaluation result

agent_id
string
required
card_id
string

Card the traces were evaluated against.

total_traces
integer

Number of traces returned by the evaluation.

results
object[]

Per-trace verdicts (or {skipped:true} for traces with no tool calls).

summary
object
message
string

Present only when the agent has no historical traces.