Keep a search or question on a schedule
create_monitorKeep a web search or an agent question and run it again on a schedule, each run compared with the last one that answered: new, dropped and moved results for a search, the answer's added, changed and removed values for a question. Use it when the person asks to keep watching, monitoring or tracking something on the web -- a brand's mentions, a ranking, a competitor's prices. To watch a whole site's pages use create_project; to ask once use search_web or run_agent. Inputs: kind 'search' with query (news=true for news results), or kind 'agent' with prompt and optional urls, schema and allowed_domains (omitted, it stays on the sites of urls, as run_agent does). schedule is hourly, daily or weekly. follow_rows=true follows a list answer record by record (read with get_monitor_rows). Cost: each run is charged as the search or agent run it is -- a results page 1 to 4 credits, an agent run up to max_credits -- and the first run starts now. max_credits_per_month skips runs once the month has spent that much. Effect: it answers with the monitor; list_monitors shows its runs. pause_monitor stops it. Needs write access; a read-only connection is refused with code 'read_only'.
Tool call template
Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {
"kind": "<kind>",
"schedule": "<schedule>"
},
"name": "create_monitor"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| kind | Yes | 'search' keeps a web search; 'agent' keeps a research question. | |
| name | No | A display name; omit for the query or the prompt. | |
| news | No | kind 'search' only: true searches news results; default false. | |
| urls | No | kind 'agent' only: up to 20 absolute URLs it starts from. | |
| query | No | kind 'search' only: what to search for, as typed into a search engine. | |
| prompt | No | kind 'agent' only: the question, 1 to 10,000 characters. | |
| schema | No | kind 'agent' only: a JSON Schema (type object or array) for the answer. | |
| schedule | Yes | How often it runs again. | |
| follow_rows | No | kind 'agent' only: true follows a list answer record by record; default false. | |
| max_credits | No | kind 'agent' only: the most one run may spend, 1 to 100,000; omit for 2,000. | |
| allowed_domains | No | kind 'agent' only: sites it stays within; omitted, the sites of urls; [] any site. | |
| max_credits_per_month | No | The most its runs may spend in a calendar month, in credits; omit or 0 for no cap. |
Output Schema
| Name | Required | Description | Default |
|---|---|---|---|
| id | No | The monitor's id, for list_monitors, pause_monitor and get_monitor_rows. | |
| code | No | Only on a failure: a machine-readable reason, e.g. read_only, validation, plan_limit or not_found. | |
| kind | No | 'search' or 'agent'. | |
| name | No | Its name; the query or the prompt unless one was given. | |
| note | No | What happens next, or what was left out. | |
| rows | No | When it follows its answer as records: {path, key, confirmAfter}. | |
| error | No | Only when the call failed: what went wrong, in words an assistant can act on. | |
| budget | No | {perRun, perMonth} in credits; 0 is no cap. | |
| detail | No | Only on a failure: the API's own wording, when the error was rephrased. | |
| status | No | 'active' or 'paused'. On an error, the HTTP status code. | |
| lastRun | No | Its latest run: {status, changed, summary, credits, refId}; refId is the search or agent run it made. | |
| request | No | The search or agent request it runs, in the API's names. | |
| schedule | No | 'hourly', 'daily' or 'weekly'. | |
| nextRunAt | No | When it runs next, ISO 8601; null while paused. | |
| spentThisMonth | No | What its runs cost this calendar month (UTC), in credits. |