/desearch/monitorsA search that runs on a schedule (`interval`, 1 hour to 7 days) and keeps only what is new: each run searches its `sources` from where the previous run ended, drops what the monitor has already seen and screens the rest for relevance with an AI classifier. New items go to your `webhook` and stay readable at `/desearch/monitors/{monitor_id}/items`. Creating a monitor is free. Each run pays for its searches at their list price plus screening per item (see `max_run_cost_usd`, the most one run can cost, held on your balance while it runs). With a webhook, the response carries `webhook_secret` once: keep it to verify the `Desearch-Signature` header.
Base URL
https://api.desearch.aiUse Create a monitor when your application needs search-grounded AI results without maintaining its own crawler, social-data pipeline, or retrieval infrastructure. The endpoint returns structured data that can be consumed by agents, research tools, dashboards, and retrieval-augmented generation workflows. Keep the request focused on the minimum fields your workflow needs so responses remain fast and easy to validate.
Send requests to POST /desearch/monitors with your Desearch API key in the authorization header. Treat the key as a server-side secret, validate all required parameters before sending the request, and handle non-2xx responses explicitly. Production integrations should use sensible timeouts, retry only transient failures with backoff, and record the response status and request identifier for troubleshooting.
Test the example request below before integrating it into a larger workflow. Confirm that the returned fields, links, and timestamps meet your freshness requirements, then add schema validation at the application boundary. For agentic systems, preserve source URLs with the generated answer so users can inspect evidence and your application can avoid presenting unsupported claims.
querystringrequiredWhat to watch, in plain words. Each source is searched with it.
namestringA label for the console and webhooks.
criteriaarray of stringsUp to 5 conditions an item must meet to be kept (each ≤ 300 characters).
One item per line, or a JSON array. Sent as an array.
sourcesarrayWhere to look.
intervalstringHow often it runs: 1h, 3h, 6h, 12h, 1d or 7d.
Default: 1d
optionsobject (JSON)Per-source settings.
A JSON object, sent as JSON. Leave it empty to omit the field. Keys: web, x, reddit, youtube, threads, tiktok, instagram, facebook, meta_ads. Allowed values: web.count: 10, 15, 20; tiktok.count: 10, 20, 30; meta_ads.count: 10, 20, 30.
min_scoreintegerRelevance (0–100) an item needs to be kept.
Default: 60
summarybooleanWrite a short digest of the items each run keeps (LLM usage + 50%).
Default: false
webhookobject (JSON)Where to send new items.
A JSON object, sent as JSON. Leave it empty to omit the field. Keys: url, events.
metadataobject (JSON)Your labels, echoed in webhooks.
A JSON object, sent as JSON. Leave it empty to omit the field.
run_nowbooleanQueue the first run at once.
Default: true
Your API key is sent in the Authorization header.
curl -X POST "https://api.desearch.ai/desearch/monitors" \
-H "Authorization: YOUR_API_KEY" \
-H "Content-Type: application/json"{
"monitor_id": "Monitor Id",
"name": "string",
"query": "Query",
"criteria": [
"string"
],
"sources": [
"string"
],
"interval": "Interval",
"options": {},
"min_score": 0,
"summary": true,
"status": "active",
"paused_reason": "user",
"webhook": {
"url": "Url",
"events": [
"string"
],
"consecutive_failures": 0,
"last_delivery": {
"event": "Event",
"status": "pending",
"attempts": 0,
"status_code": 0,
"error": "string",
"created_at": "string",
"next_attempt_at": "string",
"delivered_at": "string"
}
},
"metadata": {},
"max_run_cost_usd": 0,
"next_run_at": "string",
"last_run": {
"run_id": "Run Id",
"status": "Status",
"item_count": 0,
"cost_usd": 0,
"fail_reason": "string",
"created_at": "string",
"finished_at": "string"
},
"stats": {
"runs": 0,
"items": 0,
"spent_usd": 0
},
"created_at": "Created At",
"updated_at": "Updated At",
"webhook_secret": "string"
}Possible status codes: