Airtop API

Name: Airtop API API
Brand: Airtop API
Availability: InStock

★ Only Publicly Available SpecAI/MLMl Inferencebearer39 EndpointsREST

For Agents

Spin up cloud browser sessions, navigate pages, click and fill forms, and scrape structured data so an agent can complete real web tasks. Supports persistent profiles and async automations.

Quickstart

Get started with Airtop API in minutes using your preferred integration method.

# Add to your MCP client config (Claude Desktop, Cursor, Windsurf)
{
  "jentic": {
    "url": "https://api.jentic.com/mcp",
    "auth": "oauth"
  }
}

# Then ask your agent:
"open a cloud browser session and scrape a page"

# → Jentic returns the GET /events tool with parameter schema, agent executes.

Capabilities

What an agent can do with Airtop API API.

Create cloud browser sessions and open multiple windows for parallel page automation

Click, hover, type, and fill forms on live web pages with form-filler automations

Run page query and scrape operations to extract structured content from rendered pages

Persist browser profiles across sessions so an agent keeps login state and cookies

GET STARTED

Start building with Airtop API API

Explore with Jentic

View OpenAPI Spec

Jentic publishes the only available OpenAPI specification for Airtop API, keeping it validated and agent-ready.

Jentic publishes the only available OpenAPI specification for Airtop API, keeping it validated and agent-ready. Airtop provides cloud browser sessions purpose-built for AI agents, exposing primitives for creating sessions, opening windows, navigating pages, clicking, hovering, filling forms, and running scrape and monitor jobs. The 39-endpoint surface includes synchronous and asynchronous automation flows, a form-filler builder, page query and scrape operations, and profile persistence so a long-running agent can keep cookies and login state across runs. Authentication is bearer-token based, with operations grouped around sessions, windows, automations, and profiles.

Use Cases

Patterns agents use Airtop API API for, with concrete tasks.

★ Agent web research with persistent login

Run an AI agent that browses authenticated dashboards, internal tools, or paywalled sites. Airtop sessions and saved profiles let the agent log in once, persist cookies, and reuse the profile on later runs without re-authenticating. Each window is independent, so an agent can investigate several tabs in parallel.

Create a session with profile 'research-bot', open a window, load https://internal.example.com/dashboard, and scrape the active alerts table

Form filling and data submission

Automate multi-step web forms (lead intake, vendor portals, government applications) by creating a form-filler automation, mapping each field to source data, and executing the fill. Airtop supports both synchronous fills for short forms and asynchronous jobs for longer flows that span page transitions.

Create a form-filler automation for the URL https://supplier.example.com/onboarding, map company name, address, and tax ID, and execute the fill synchronously

Targeted web scraping with page queries

Extract structured data from sites that render content client-side. Airtop loads the page in a real browser, waits for JavaScript to settle, and lets the agent issue page queries or scrape calls to pull specific elements. Use sessions to keep performance up across many pages on the same domain.

Open a window at https://store.example.com/products, scrape product names and prices into JSON

AI agent integration via Jentic

Use Airtop as the browser layer for an AI agent built on Jentic. The agent searches for browsing operations, loads the Airtop schema, and calls session and window endpoints without managing browser infrastructure. Credentials stay in the Jentic vault so the bearer token never enters the agent context.

Search Jentic for 'open a browser session and scrape a page', load the matched Airtop operation, and execute it for https://news.example.com

Key Endpoints

39 endpoints — jentic publishes the only available openapi specification for airtop api, keeping it validated and agent-ready.

METHOD

PATH

DESCRIPTION

POST

/v1/sessions

Create a new cloud browser session

GET

/v1/sessions

List active sessions

POST

/v1/sessions/{sessionId}/windows

Open a new window in a session

POST

/v1/sessions/{sessionId}/windows/{windowId}/click

Click an element on the page

POST

/v1/sessions/{sessionId}/windows/{windowId}/fill-form

Fill a form synchronously

POST

/v1/sessions/{sessionId}/windows/{windowId}/monitor

Monitor a window for a condition

PUT

/v1/sessions/{sessionId}/save-profile-on-termination/{profileName}

Save profile on session end

POST

/v1/sessions

Create a new cloud browser session

GET

/v1/sessions

List active sessions

POST

/v1/sessions/{sessionId}/windows

Open a new window in a session

POST

/v1/sessions/{sessionId}/windows/{windowId}/click

Click an element on the page

POST

/v1/sessions/{sessionId}/windows/{windowId}/fill-form

Fill a form synchronously

Why though Jentic?

Three things that make agents converge on Jentic-routed access.

Credential isolation

Airtop bearer tokens are stored encrypted in the Jentic vault. Agents receive a scoped execution capability — the raw token never enters the agent's prompt or memory.

Intent-based discovery

Agents search by intent (e.g., 'open a browser session and click a button') and Jentic returns the matching Airtop operation with its input schema, so the agent calls the correct endpoint without browsing docs.

Time to first call

Direct Airtop integration: 1-2 days to wire bearer auth, session lifecycle, and async polling. Through Jentic: under an hour — search, load schema, execute.

Related APIs

Alternatives and complements available in the Jentic catalogue.

Alternative

Browserless

Headless Chrome as a service — lower-level browser automation without Airtop's agent-oriented primitives.

Choose Browserless when you want raw Puppeteer/Playwright control over the browser. Choose Airtop when you want higher-level agent actions like 'fill this form' or 'scrape this page' without scripting.

Alternative

Apify

Web scraping platform with prebuilt actors and a managed browser pool.

Choose Apify when you want prebuilt scrapers for known sites. Choose Airtop when the agent needs to drive an arbitrary site interactively.

Complementary

OpenAI API

Use OpenAI to plan agent steps and parse Airtop's scraped output.

Pair OpenAI with Airtop when the agent needs an LLM to decide what to click next or to extract structured data from the rendered page.

FAQs

Specific to using Airtop API API through Jentic.

Why is there no official OpenAPI spec for Airtop API?

Airtop does not publish an OpenAPI specification. Jentic generates and maintains this spec so that AI agents and developers can call Airtop API via structured tooling. It is validated against the live API and kept up to date. Get started at https://app.jentic.com/sign-up.

What authentication does the Airtop API use?

Airtop uses HTTP bearer token authentication — every request must include an Authorization: Bearer <token> header. Through Jentic, the bearer token is stored in the encrypted Jentic vault and exchanged at execution time, so the raw token never enters the agent context.

Can I keep a browser logged in across agent runs with Airtop?

Yes. Call PUT /v1/sessions/{sessionId}/save-profile-on-termination/{profileName} to save the browser profile when the session ends, then create new sessions referencing the same profile name to restore cookies and storage. This avoids re-authenticating on each run.

What is the difference between sync and async automations?

Synchronous endpoints under /v1/sessions/{id}/windows/{id}/* return when the action completes (good for short clicks and fills). Async endpoints under /v1/async/sessions/{id}/windows/{id}/* enqueue the work and return a job id for long-running flows like multi-step form fillers and large scrapes.

How do I scrape a page with Airtop through Jentic?

Use the Jentic search query 'open a browser session and scrape a page'. Jentic will return the matching Airtop operations (POST /v1/sessions, POST /v1/sessions/{sessionId}/windows, and the scrape endpoint). Load the operation schema and execute — Jentic handles credential injection.

Are there rate limits on the Airtop API?

The OpenAPI spec does not declare explicit rate limits; quotas are tied to the customer plan. Treat sessions as a finite resource, end them when work is done, and watch for 429 responses on burst traffic.