For the complete documentation index, see llms.txt. This page is also available as Markdown.

Health checks

Get Health Check Status

get

The settings, the run in progress, and the latest finished run by provider and by offering.

Authorizations
Otari-KeystringRequired

Enter your API key here (sent as Otari-Key header).

Responses
200

Successful Response

application/json

The Health page: settings, the run in progress, and the latest finished run by provider.

get/api/v1/health-checks
GET /api/v1/health-checks HTTP/1.1
Otari-Key: YOUR_API_KEY
Accept: */*
200

Successful Response

{
  "last_run": {
    "failed": 1,
    "finished_at": "2026-01-01T00:00:00.000Z",
    "id": "text",
    "included_reasoning": true,
    "passed": 1,
    "started_at": "2026-01-01T00:00:00.000Z",
    "total": 1
  },
  "providers": [
    {
      "answered": 1,
      "history": [
        {
          "answered": 1,
          "started_at": "2026-01-01T00:00:00.000Z",
          "total": 1
        }
      ],
      "provider": "text",
      "status": "ok",
      "top_error": "text",
      "top_error_count": 1,
      "total": 1
    }
  ],
  "results": [
    {
      "checked_at": "2026-01-01T00:00:00.000Z",
      "error": "text",
      "failed_runs": 0,
      "failing_since": "2026-01-01T00:00:00.000Z",
      "latency_ms": 1,
      "provider": "text",
      "selector": "text",
      "status": "pass",
      "status_code": 1
    }
  ],
  "running": {
    "failed": 1,
    "finished_at": "2026-01-01T00:00:00.000Z",
    "id": "text",
    "included_reasoning": true,
    "passed": 1,
    "started_at": "2026-01-01T00:00:00.000Z",
    "total": 1
  },
  "settings": {
    "alert_emails": [
      "text"
    ],
    "interval_minutes": 1,
    "max_tokens": 1,
    "next_run_at": "2026-01-01T00:00:00.000Z",
    "prompt": "text",
    "reasoning_every_runs": 1,
    "reasoning_max_tokens": 1,
    "run_requested": true,
    "schedule_enabled": true
  }
}

Disable Failing Models

post

Stop serving every model the latest run found failing, on its hosted provider.

Each stays listed on the Hosted providers page, priced, to be switched back on there.

Authorizations
Otari-KeystringRequired

Enter your API key here (sent as Otari-Key header).

Responses
200

Successful Response

application/json

What switching off the failing models did.

disabledintegerRequired

How many models were switched off on their hosted provider.

post/api/v1/health-checks/disable-failing
POST /api/v1/health-checks/disable-failing HTTP/1.1
Otari-Key: YOUR_API_KEY
Accept: */*
200

Successful Response

{
  "disabled": 1
}

Run Health Checks

post

Start a run within a minute, whether or not the schedule is on.

Authorizations
Otari-KeystringRequired

Enter your API key here (sent as Otari-Key header).

Responses
202

Successful Response

application/json

Whether the checks run, how often, and what they send.

alert_emailsstring[]Required
interval_minutesintegerRequired
max_tokensintegerRequired
next_run_atstring · date-time · nullableRequired
promptstringRequired
reasoning_every_runsintegerRequired
reasoning_max_tokensintegerRequired
run_requestedbooleanRequired
schedule_enabledbooleanRequired
post/api/v1/health-checks/run
POST /api/v1/health-checks/run HTTP/1.1
Otari-Key: YOUR_API_KEY
Accept: */*
202

Successful Response

{
  "alert_emails": [
    "text"
  ],
  "interval_minutes": 1,
  "max_tokens": 1,
  "next_run_at": "2026-01-01T00:00:00.000Z",
  "prompt": "text",
  "reasoning_every_runs": 1,
  "reasoning_max_tokens": 1,
  "run_requested": true,
  "schedule_enabled": true
}

Get Health Check Settings

get

Whether the checks run on a schedule, how often, and what they send.

Authorizations
Otari-KeystringRequired

Enter your API key here (sent as Otari-Key header).

Responses
200

Successful Response

application/json

Whether the checks run, how often, and what they send.

alert_emailsstring[]Required
interval_minutesintegerRequired
max_tokensintegerRequired
next_run_atstring · date-time · nullableRequired
promptstringRequired
reasoning_every_runsintegerRequired
reasoning_max_tokensintegerRequired
run_requestedbooleanRequired
schedule_enabledbooleanRequired
get/api/v1/health-checks/settings
GET /api/v1/health-checks/settings HTTP/1.1
Otari-Key: YOUR_API_KEY
Accept: */*
200

Successful Response

{
  "alert_emails": [
    "text"
  ],
  "interval_minutes": 1,
  "max_tokens": 1,
  "next_run_at": "2026-01-01T00:00:00.000Z",
  "prompt": "text",
  "reasoning_every_runs": 1,
  "reasoning_max_tokens": 1,
  "run_requested": true,
  "schedule_enabled": true
}

Update Health Check Settings

patch

Change the settings. Turning the schedule on starts a run within a minute.

Authorizations
Otari-KeystringRequired

Enter your API key here (sent as Otari-Key header).

Body

Settings to change. An omitted field keeps its value.

alert_emailsstring · email[] · max: 20 · nullableOptional

Who is emailed when a model stops answering and when it answers again. An empty list emails nobody.

interval_minutesinteger · min: 15 · max: 1440 · nullableOptional

Minutes between scheduled runs.

max_tokensinteger · min: 1 · max: 4096 · nullableOptional

Output cap for a model that does not reason.

promptstring · min: 1 · max: 2000 · nullableOptional

The user message every model is sent. A check passes on any answer, whatever it says.

reasoning_every_runsinteger · max: 48 · nullableOptional

Include reasoning models every Nth run; 0 leaves them out.

reasoning_max_tokensinteger · min: 1 · max: 8192 · nullableOptional

Output cap for a reasoning model.

schedule_enabledboolean · nullableOptional

Run the checks on a schedule.

Responses
200

Successful Response

application/json

Whether the checks run, how often, and what they send.

alert_emailsstring[]Required
interval_minutesintegerRequired
max_tokensintegerRequired
next_run_atstring · date-time · nullableRequired
promptstringRequired
reasoning_every_runsintegerRequired
reasoning_max_tokensintegerRequired
run_requestedbooleanRequired
schedule_enabledbooleanRequired
patch/api/v1/health-checks/settings
PATCH /api/v1/health-checks/settings HTTP/1.1
Otari-Key: YOUR_API_KEY
Content-Type: application/json
Accept: */*
Content-Length: 47

{
  "interval_minutes": 60,
  "schedule_enabled": true
}
{
  "alert_emails": [
    "text"
  ],
  "interval_minutes": 1,
  "max_tokens": 1,
  "next_run_at": "2026-01-01T00:00:00.000Z",
  "prompt": "text",
  "reasoning_every_runs": 1,
  "reasoning_max_tokens": 1,
  "run_requested": true,
  "schedule_enabled": true
}

Last updated

Was this helpful?