Sign in
assistants

Return the live reachability of the assistant's runtime pod.

The web frontend polls this endpoint to distinguish a pod that is waking from idle-sleep (transient) from one that is genuinely unreachable (surface an error).

post/v1/assistants/{id}/connection-status/

Request

  • The document declares no server URL.
  • Auth: one of:
    • HTTP bearer
    • API key in cookie sessionid
    • API key in header X-Session-Token
    • API key in header Authorization

Path parameters

idstring required

Headers

Vellum-Organization-Idstring uuid

Required if using Cookie-based or X-Session-Token authentication methods.

Response

state'ready' | 'waking' | 'crash_loop' | 'not_found' | 'unreachable' required
  • ready - ready
  • waking - waking
  • crash_loop - crash_loop
  • not_found - not_found
  • unreachable - unreachable
is_awakeboolean required
pod_statusstring nullable required
waking_sincestring date-time nullable required
last_ready_atstring date-time nullable required
crash_loop_sincestring date-time nullable required
detailstring nullable required
pod_error_kindstring nullable

Classified root cause of a CRASH_LOOP, when vembda was able to infer one (e.g. 'out_of_storage' when the pod has run out of disk space). Clients can branch on this to show targeted remediation UI such as a link to the storage doctor.

Changes

No recorded changes to this operation across all 50 revisions of this API.