/v1/monitors

Schedule recurring /v1/ask-shaped prompts. The hourly cron dispatches every monitor whose next_run_at has passed. Results are archived just like any other /v1/ask call — retrievable for 30 days — and can fire an ask.completed webhook per run.

These endpoints are session-authenticated via dashboard login, not via bearer token.

Legacy endpoint. This documents an older surface kept for back-compat with existing integrations. New integrations should use /v1/watch instead — see the quickstart for the modern API.

Monitor object

Field	Type	Description
`id`optional	`string (uuid)`	Monitor identifier.
`account_id`optional	`string`	Your account UUID.
`name`optional	`string`	Human label. 1–120 chars.
`prompt`optional	`string`	The prompt to run on every tick. 1–10,000 chars.
`providers`optional	`('openai' \| 'anthropic' \| 'gemini' \| 'perplexity')[]`	1–4 entries.
`cadence`optional	`'hourly' \| 'daily' \| 'weekly'`	Run frequency.
`track_brands`optional	`string[]`	Up to 20 brand names to extract on every run.
`webhook_id`optional	`string (uuid) \| null`	Registered webhook_endpoints.id. If set, each run fires an ask.completed delivery.
`enabled`optional	`boolean`	Paused monitors do not run.
`next_run_at`optional	`string (ISO) \| null`	When the next tick is due.
`last_run_at`optional	`string (ISO) \| null`	When the last run finished.
`created_at`optional	`string (ISO)`
`updated_at`optional	`string (ISO)`

GET /v1/monitors

GET/v1/monitors— List all monitors for the calling account

Response

json

{
  "request_id": "…",
  "monitors": [ /* Monitor[] */ ]
}

POST /v1/monitors

POST/v1/monitors— Create a new monitor

Request body

Field	Type	Description
`name` required	`string`	1–120 chars.
`prompt` required	`string`	1–10,000 chars.
`providers` required	`('openai' \| 'anthropic' \| 'gemini' \| 'perplexity')[]`	1–4 entries.
`cadence`optional	`'hourly' \| 'daily' \| 'weekly'`	Default: `'daily'`
`track_brands`optional	`string[]`	Up to 20 entries, each 1–80 chars.
`webhook_id`optional	`string (uuid)`	A registered webhook endpoint to deliver ask.completed events to.
`enabled`optional	`boolean`	Default: `true`

GET /v1/monitors/:id

GET/v1/monitors/:id— Fetch one monitor and its 10 most recent runs

Response

json

{
  "request_id": "…",
  "monitor": { /* Monitor */ },
  "runs": [
    {
      "id": "…",
      "monitor_id": "…",
      "ran_at": "2026-04-25T08:00:00.000Z",
      "status": "succeeded",
      "diff_summary": "2 new mentions vs previous run",
      "response": { /* NormalizedResponse */ },
      "cost_cents": 75
    }
  ]
}

PATCH /v1/monitors/:id

PATCH/v1/monitors/:id— Update any subset of fields

Changing cadence re-anchors next_run_at relative to now — a user switching from weekly → hourly doesn't have to wait a week for the next tick.

DELETE /v1/monitors/:id

DELETE/v1/monitors/:id— Permanently delete the monitor

json

{ "request_id": "…", "deleted": true }

Pricing

Each monitor run is billed as a normal /v1/ask call using the pricing tiers in Billing. Runs use cache_scope: "private" so repeat prompts with identical inputs still return cached at 2¢.