The Monitoring API watches resources of other HasData APIs on a schedule and tells you what changed. A monitor is one API endpoint, one or more resources (requests to it, such as an Amazon ASIN or a Zillow search), the fields of the response to compare, a schedule, and where to deliver changes.Every check calls the API for each resource, keeps a snapshot of the tracked fields and diffs it against the previous one. Changes land in a feed you can poll, in a webhook, or in your inbox.Monitors can watch most HasData APIs: search results, Maps, Amazon, Walmart, Zillow, social profiles, job boards, Web Scraping and more. GET /monitors/endpoints lists them with their ids, the params each accepts and the fields worth tracking. An endpoint id is the API’s name with underscores, such as amazon_product, google_serp or zillow_listing. Chat APIs are left out: their answers differ on every call.In the dashboard, open an API in the Playground, set up a request and choose Monitor this request.
Create a Monitor
The first check runs as soon as the monitor is created and becomes the starting point for comparisons.
An id from GET /monitors/endpoints, such as amazon_product, google_serp or web_scraping.
resources
yes
One request per resource: { "params": { … } }, the same params the API itself takes.
fields
no
What to compare. Omitted, the endpoint’s default fields; [] compares the whole response. See Fields to Track.
sharedParams
no
Params set once for every resource. For Web Scraping, aiExtractRules.
schedule
no
{ "preset": "5m" | "15m" | "1h" | "6h" | "24h" | "7d" } or { "cron": "0 9 * * *", "timezone": "Europe/Berlin" }. 24h by default; 5 minutes is the minimum.
delivery
no
webhook: { url, secret? } and email: { to: [ … ] }; see Notifications.
goal
no
What matters, in plain words. With it, every change is judged by AI; see Notifications.
name
no
Without it the first resource names the monitor.
Request Cost and API Credits
A check costs what the underlying API costs: every resource is one request to it, charged only when it succeeds. A monitor of two Amazon products checked every 6 hours spends 2 × 10 credits per check, 80 credits a day. The AI judge, when a goal is set, adds 2 credits per change it looks at.
Run one request through the API first to see its exact cost; GET /monitors/endpoints gives the typical credits per request for each endpoint.
When the balance runs out, the monitor pauses itself (pauseReason: "no_credits") and resumes when you call POST /monitors/{id}/resume.
Next Steps
Fields to Track: what a check compares, and how lists are read.
Checks and Changes: running a check now, polling it, reading the changes feed.