Server telemetry (optional)

Counters kept by the sample AI server when telemetry is enabled. Also served at /api/telemetry.

GET /api/response/telemetry

cURL

curl --request GET \
  --url 'https://api.example.com/api/response/telemetry' \
  --header 'x-widget-token: YOUR_API_KEY'

200

{
  "ok": true,
  "generatedAt": "2026-08-31T14:22:05Z",
  "telemetry": {
    "requests": 1.5,
    "errors": 1.5,
    "avgLatencyMs": 250,
    "lastLatencyMs": 250,
    "kbHits": 1.5,
    "kbMisses": 1.5,
    "last": {
      "provider": "string",
      "model": "string",
      "stream": true
    }
  }
}

Counters kept by the sample AI server when telemetry is enabled. Also served at /api/telemetry.

Response

Telemetry

ok · boolean

generatedAt · string (date-time)

telemetry · object

Show telemetry

requests · number

errors · number

avgLatencyMs · number

lastLatencyMs · number

kbHits · number

kbMisses · number

last · object

Show last

provider · string

model · string

stream · boolean

Status codes

Code	Meaning
200	OK
401	Error

cURL

curl --request GET \
  --url 'https://api.example.com/api/response/telemetry' \
  --header 'x-widget-token: YOUR_API_KEY'

200

{
  "ok": true,
  "generatedAt": "2026-08-31T14:22:05Z",
  "telemetry": {
    "requests": 1.5,
    "errors": 1.5,
    "avgLatencyMs": 250,
    "lastLatencyMs": 250,
    "kbHits": 1.5,
    "kbMisses": 1.5,
    "last": {
      "provider": "string",
      "model": "string",
      "stream": true
    }
  }
}