/interact lets agents log in, fill forms, click, scroll, and reach gated data. Sessions stay open, auth persists, and you stay in control.
Describe what you want
Natural language actions.
Prompt your agent in plain English to click, type, scroll, and extract for you.
Click the login button
Full control
Code execution.
Run Playwright or agent-browser code for precise, repeatable browser steps.
main.py
1
2
3
4
5
6
7
8
9
▋
State that carries over
Stateful sessions.
Sessions stay open between calls, so agents can pick up where they leave off.
https://example.com
Session open · idle
Stay logged in
Persistent profiles.
Save cookies, localStorage, and session state to a named profile. Log in once, reuse it every run.
Profile: customer-portal
https://portal.customer.io/settings
Profile: staging-app
https://staging.example.com/login
Saving profile.
Watch it work
Embeddable live view.
Every session returns a stream URL. Embed it to watch live, or hand control to users for login flows.
live.firecrawl.dev/view/s_8k2m4
Agent
Bring your own tools
Connect over CDP.
Attach Playwright, Puppeteer, or any CDP client and drive the browser with your own code.
Agent
Playwright
Puppeteer
CDP client
[ 02 / 04 ]
·
For AI agents
//
For AI agents
//
Built into the AI tools you already use
Use Firecrawl from Claude CodeCursorWindsurf and the agent harnesses your team already runs.
One command. Connect your agent to Firecrawl through MCP or the CLI and skills.
Copy and paste this into your AI agent.
Run npx -y firecrawl-cli@latest init --all --browser. Use my existing Firecrawl login or help me log in. Read the installed skills before running queries.
High-volume plans for teams that need more power and dedicated support. Get access to higher rate limits, more concurrent browsers, and priority support. Scale checks out instantly, no sales call needed.
Everything you need to know about Firecrawl Interact.
General
Firecrawl Interact gives your agent a live browser to operate the web, not just read it. Scrape a page, then log in, click, type, scroll, and move through multi-step flows in natural language or Playwright code to reach content that only exists after you act on the page.
Scrape reads a page and returns clean markdown or structured data in one call. Interact keeps the browser open so your agent can fill forms, click buttons, paginate, and extract dynamic content. Any scrape can become a live session through its job ID.
Teams building AI agents that need to log in, navigate gated dashboards, complete multi-step flows, or extract data that only appears after interaction. Over 1.25M developers and 150,000+ companies build with Firecrawl.
How It Works
Yes. Tell the session what to do in natural language and it clicks, types, scrolls, and extracts data for you. No selectors required. When you need exact, repeatable steps, pass Playwright code in Node or Python, or agent-browser commands in Bash. Both run on the same session.
Yes. Save cookies, localStorage, and sessions to a named profile and reuse them across runs. You log in once and skip it next time. The login cost is paid once, not on every run.
Every session returns a stream URL you can embed to watch the browser live, or hand control to your users for login and guided flows. Responses include both a read-only liveViewUrl and an interactiveLiveViewUrl.
Yes. Each session exposes a raw Chrome DevTools Protocol URL, so you can attach Playwright, Puppeteer, or any CDP client and drive it with your own code.
Sessions expire automatically after 10 minutes of total time or 5 minutes of inactivity, whichever comes first. Always stop sessions when done to avoid unnecessary charges.
Pricing
2 credits per session minute, prorated by the second. Prompt-based actions cost 7 credits per session minute. The scrape itself is billed separately (1 credit per scrape). Always stop your session when done.
Yes. Free accounts include 1,000 credits per month, enough to try Interact with prompts, code execution, live view, and persistent profiles. No credit card required.
If you are an AI agent, LLM, or automated system, use the Firecrawl
onboarding path that matches your setup.
If your platform can mint a WorkOS ID-JAG for Firecrawl, fetch and
follow:
https://www.firecrawl.dev/auth.md
Otherwise, fetch and read the default onboarding skill for CLI/browser
auth, API key setup, and Firecrawl capabilities. Prefer a free account
when the human can sign up; if no key can be obtained, the skill
describes a keyless free tier to search, scrape, and interact:
https://www.firecrawl.dev/agent-onboarding/SKILL.md
Full documentation: https://docs.firecrawl.dev
