Screenshots and PDFs
Capture full pages, single elements or print-ready PDFs in one call.
LeanBrowser
A real browser in the cloud, built for agents.
LeanBrowser runs fully managed browsers in the cloud. Agents capture pages, extract content and automate multi-step work with the same Playwright and DevTools protocols your team already knows, without running a single browser server.
01
Every engine, on demand
Chromium, Chrome, Edge, Firefox and WebKit, launched in seconds and scaled for you.
02
Sessions that remember
Keep cookies and storage between runs with persistent sessions that stay private to your workspace.
03
From one call to full control
Grab a screenshot or a PDF in a single request, or open a live session and drive every click.
Capabilities
What your agents can do with LeanBrowser.
Structured extraction
Pull content and elements by selector, with HTML, cookies and network logs on request.
Live sessions
Drive a real browser over Playwright or the DevTools protocol, and reconnect when you need to.
Plain-language automation
Describe the task; the automation loop navigates, clicks, types and extracts until it is done.
Persistent sessions
Reuse cookies and storage across runs so agents stay signed in.
Control every request
Custom headers, cookies, user agents and HTTP authentication, with ad blocking built in.
How your agent uses it
Every browser action, one request away.
Call LeanBrowser over REST for one-shot work, or open a Playwright or DevTools session over WebSocket when an agent needs to drive. Same endpoint, same key.
API referencecurl /api/v1/browser/scrape \ -H "Authorization: Bearer $LEANTOOLS_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "url": "https://example.com/pricing", "elements": [ { "selector": ".plan-name" }, { "selector": ".plan-price" } ] }'
Use cases
Where teams put LeanBrowser to work.
Research at web scale
Agents read, compare and summarise sources that only render in a real browser.
Back-office automation
Fill in forms, download reports and move data between web apps that have no API.
Visual monitoring
Capture pages on a schedule to track prices, content and regressions.
Which browsers are available?
Chromium, Chrome and Edge for one-shot calls and DevTools sessions, plus Firefox and WebKit through Playwright.
Do I need my own model for automation?
Plain-language automation runs on an OpenAI-compatible model of your choice: pass its key or endpoint with the task. Screenshots, extraction and live sessions need nothing extra.
Are sessions private?
Yes. Sessions and saved browser state belong to your workspace and are never shared.
How many sessions can run at once?
Every workspace can run several live sessions in parallel. Talk to us if your agents need more.
Better together
Pairs well with LeanBrowser.
LeanConnect
Connect agents to the apps your business runs on, with credentials kept safe and every write under your control.
LeanMachine
Ephemeral Linux sandboxes for code, builds and file work. Ready in seconds, gone when the job is done.
LeanMemory
A memory that learns from use: structured, unstructured and event data in one place, with recall that gets more accurate the more your agents work.
Start building with LeanBrowser.
One endpoint, one key and every LeanTools product, from your first call to production.