Gets the agent's relevance-gate decision counts for the operator overview card.
Returns responded / silenced / budget-bypassed counts (and a silence rate) over the requested origin + window. Null/null window covers all time; otherwise it defaults to 30 days before to, mirroring the cost-timeseries windowing so the surfaces reconcile.
Authorization
Bearer JWT Authorization header using the Bearer scheme. Enter 'Bearer' [space] and then your token.
In: header
Path Parameters
The unique identifier of the agent.
The tenant identifier
Query Parameters
Execution origin to roll up. Defaults to PRODUCTION.
"PRODUCTION""UNSPECIFIED" | "PRODUCTION" | "TEST"Optional window start as Unix milliseconds (inclusive).
int64Optional window end as Unix milliseconds (exclusive). Defaults to now when a window is requested.
int64Optional environment slug to scope the rollup to.
Response Body
curl -X GET "https://api.fruxon.com/v1/tenants/string/agents/string/relevanceActivity"{
"responded": 0,
"silenced": 0,
"bypassed": 0,
"cost": 0,
"evaluated": 0,
"total": 0,
"silenceRate": null
}{
"type": "string",
"title": "string",
"status": 0,
"detail": "string",
"instance": "string",
"property1": null,
"property2": null
}Lists the agent's revised, failed or unchecked runs, newest first.
The drill-down behind `judgeActivity.revised` (the default), `failed` or `errors`, chosen by outcome, over the same origin, window and environment. Each row carries the participant and the channel address (or, when the session recorded none, the topic) so the client can open the conversation at the accepted reply, whose messages carry this `executionId` as `executionRecordId`. Each row also carries `reasons`, the metrics the judge took points off with its shortened, redacted reason: for a reply that passed after a rewrite, why the draft was sent back. The rejected drafts themselves are in the run's trace.
Gets how much work the agent's hand-offs were for the team, bucketed over time.
A hand-off is counted once it closed and the team's exchange with the customer is on record; until then it is `pending`. Each is split three ways: `withoutAsking` — the team answered and finished without needing anything from the customer; `asked` — the customer wrote between the team's first and last message, so the team waited on an answer; `noReply` — nobody answered in the chat. Each hand-off is counted in the bucket it opened in; buckets with none are omitted. Counting stops where the workspace's run history does, as the conversations series' does, so a window reaching further back is counted from `countedFromMs`. `measured` is false for an agent none of whose hand-offs has ever been measured — one whose escalations go where the team's replies are never seen.