Reference & Lookup Data · Integration
AgentQL
Add AgentQL to your product for your customers, and give your AI agents governed access to it.
AgentQL turns a page into JSON: you POST to https://api.agentql.com/v1/query-data with an X-API-Key header, and the request must carry either a query or a prompt but never both, and either a url or raw html but never both. A GET against the same path returns 405, which is the usual first sign that the endpoint is alive but being called wrongly. The params object controls the fetch itself with wait_for, scroll behaviour, mode, proxy configuration and an optional screenshot, and every response carries a metadata block with a request_id you should log because it is what support asks for. There is a /v1/query-document endpoint that takes PDFs and images as multipart form data with a stringified JSON body, and /v1/tetra/sessions which hands back a cdp_url for driving a remote browser directly. Reference & Lookup Data is the nearest fit, since what you get back is query results rather than any hosted records. fastn holds each customer's API key and keeps the connection maintained as keys rotate.
In your product
Embedded for your customers. Per-tenant auth, no per-customer code, maintained by fastn.
Let a customer connect their own AgentQL key so page extraction runs on their quota and under their proxy settings
Extract structured fields from a supplier or listing page a customer pastes into your product
Parse uploaded PDF invoices or spec sheets into typed JSON through the document query endpoint
Re run a saved query on a schedule and record when the extracted values on a watched page change
For your AI agents
Governed, audited access for the agents you build, through the MCP server.
Let an agent pull named fields off a URL and return typed JSON instead of scraped text
Have an agent open a remote browser session for a page that needs interaction before the data appears
Audit per tenant which URLs an agent queried and log the returned request_id for each extraction
Example prompt
Pull the price, stock status and delivery estimate from this product page.
Set up AgentQL in 4 steps
- 01Enable the AgentQL connector from your fastn dashboard.
- 02Have each customer authorise their own AgentQL account, so calls run under their credentials rather than a shared key.
- 03Decide which lookups, results and reports your product needs, map those fields, then enable the actions and triggers you want.
- 04Call it from your product and expose it to your agents through the same governed connection.
Why teams use the AgentQL integration
What you get by embedding it with fastn instead of building it yourself.
- Ship an AgentQL integration without building it. Your customers connect their own AgentQL account inside your product and work their lookups, results and reports there, with no per-customer code on your side.
- Handle the part that actually costs time: these are metered read sources, so an uncached lookup on every page view burns the customer's allowance. fastn owns the auth, token refresh, rate limits, pagination and breaking-change fixes, so an AgentQL update is not your on-call problem.
- One integration serves your product and your agents. The same governed AgentQL connection powers in-product features and gives AI agents scoped, audited access, so you answer questions from an authoritative outside source without holding that data yourself without wiring it twice.
Used by these teams
Compare with
Clearbit
APIVoidOften used alongside
Tools the same teams tend to run next to AgentQL, across other categories.
Salesforce
HubSpot
Notion
SnowflakeAgentQL integration FAQ
How do I add an AgentQL integration to my product?
Enable the AgentQL connector in your fastn dashboard, then let each customer authenticate their own AgentQL account. fastn handles the OAuth flow, token storage and refresh per tenant, so there is no AgentQL client code in your app and no per-customer branch in your codebase. Setup is 4 steps.
Do my customers each connect their own AgentQL account?
Yes. Every connection is scoped to the individual customer, so each authorises their own AgentQL account and only ever sees their own lookups, results and reports. That per-tenant isolation is the point of an embedded integration: you support the long tail of customer setups without maintaining an integration per customer.
Can AI agents use this AgentQL integration?
Yes. The same connection is exposed to your agents through the fastn MCP gateway, with permissions scoped per tenant and every call audited. Let an agent pull named fields off a URL and return typed JSON instead of scraped text
Who maintains the AgentQL integration?
fastn does. When AgentQL changes an endpoint, deprecates a field or alters its auth, the fix lands in the connector rather than in your backlog, and your customers' connections keep working.
Whose credentials does an AgentQL lookup run under?
Each customer's own. Quota and cost land on the customer that caused them rather than on a shared key you have to meter and re-bill, and one heavy customer cannot exhaust another's allowance.
How should AgentQL results be cached?
Per tenant, with a window that matches how fast the underlying data actually moves. AgentQL is a read source rather than a system of record, so the same query returns the same answer until the source changes, and caching is the difference between a sustainable integration and one that exhausts a customer's allowance on a busy page.
What can I build with the AgentQL integration?
A common starting point: let a customer connect their own AgentQL key so page extraction runs on their quota and under their proxy settings. Teams also use it for the other use cases listed above, and expose it to agents for governed reads and writes.
How much does the AgentQL integration cost?
It is included. Pricing is based on connected accounts, not on how many connectors you enable, so adding AgentQL does not change your per-connector cost. You can start free with 3 connected accounts.
Add AgentQL to your product
Start free with 3 connected accounts. No sales call required, and no per-customer integration code.