Apify
Web scraping and automationCloud platform of ready-made scrapers ("Actors") and automation. Its hosted MCP server at mcp.apify.com lets AI assistants run Actors and get structured web data back.
MCPAPI/CLIBulk import
Research
B
Browser Use
Browser automation for AI agentsIndependent review. IonBin has no partnership with Browser Use and was not paid for this page. Our evaluation is our own, based on public documentation and our research. Product names are trademarks of their owners and are used for identification only. How disclosure works.
Open-source library and cloud service that lets AI agents operate a web browser. The cloud MCP server connects Claude, Cursor, Windsurf or any MCP client. Useful when a site has no API; review anything that submits or buys.
Best for
- Automating web tasks for sites without an API
Not for
- Tasks a stable API already covers
Top use cases
- Run browser tasks from Claude Code or Cursor through the cloud MCP server source ↗
- Run browser tasks from code or the cloud API source ↗
In depth
Pricing
No subscription; credits never expire. Browsers cost $0.02 per hour and residential proxy traffic $5/GB; browser agents add model cost plus 20%. Eligible new signups get a one-time $1 credit, and top-ups start at $5. Checked 2026-10-17. source ↗
How to use it with AI
How Browser Use connects to each AI platform: native connector or MCP, API through code the agent writes, browser control, or file upload.
| Platform | MCP / connector | API (agent-written code) | Browser | Upload / export |
|---|---|---|---|---|
| Claude (incl. Claude Code) | Not documented Browser Use has an official MCP server and Claude (incl. Claude Code) supports MCP, but Browser Use's docs don't name this pairing. Auth setup may differ. Browser Use MCP docs ↗Claude (incl. Claude Code) MCP docs ↗ | Confirmed Claude (incl. Claude Code) can run code; Browser Use publishes an API. Browser Use API ↗Claude (incl. Claude Code) docs ↗ | Confirmed Claude Code with the Claude in Chrome extension (Chrome, Edge and other Chromium browsers) Claude (incl. Claude Code) browser docs ↗ | Confirmed Export from Browser Use, then works on the files in your project; chat accepts attachments Claude (incl. Claude Code) docs ↗ |
| OpenAI Codex | Not documented Browser Use has an official MCP server and OpenAI Codex supports MCP, but Browser Use's docs don't name this pairing. Auth setup may differ. Browser Use MCP docs ↗OpenAI Codex MCP docs ↗ | Confirmed OpenAI Codex can run code; Browser Use publishes an API. Browser Use API ↗OpenAI Codex docs ↗ | Confirmed Only in the ChatGPT desktop app; not in the Codex CLI or IDE extension OpenAI Codex browser docs ↗ | Confirmed Export from Browser Use, then works on local repositories OpenAI Codex docs ↗ |
| Google Antigravity | Not documented Browser Use has an official MCP server and Google Antigravity supports MCP, but Browser Use's docs don't name this pairing. Auth setup may differ. Browser Use MCP docs ↗Google Antigravity MCP docs ↗ | Confirmed Google Antigravity can run code; Browser Use publishes an API. Browser Use API ↗Google Antigravity docs ↗ | Confirmed Browser subagent via /browser, documented as an Antigravity App feature; the IDE has separate browser docs Google Antigravity browser docs ↗ | Confirmed Export from Browser Use, then works on local projects and worktrees Google Antigravity docs ↗ |
Confirmed means both the tool's and the platform's official docs describe that route (links in the cell); IonBin has not tested every pairing. Not documented means it may work, but no official docs describe that route end to end. Checked Oct 17, 2026.
Agent compatibility
Agent platform MCPAPI/CLIBulk import
| Level | Status | Detail / source |
|---|---|---|
| Native connector / MCP | Yes | Official docs ↗ |
| API / CLI | Yes | Official docs ↗ |
| Browser-only | No | Web app; also reachable programmatically |
| Bulk upload / import | Yes | Concurrent cloud sessions, from 10 and rising with spend tier (https://docs.browser-use.com/cloud/guides/concurrency.md) |
Recorded Oct 17, 2026. Setup and plan limits vary; check the vendor's docs. How levels are assigned.
Works well with
As an agent platform
MCP client support to verify.
Workflows using this tool
- Automate a web task that has no API Reviewed
Review
No long-form review yet. Verdict is based on research, not hands-on testing.