Marketplace for trusted AI agent skills

Skill Detail

Browse from MCP clients with Browser Use MCP Server

Expose browser-use web automation through an MCP server so Cursor or another MCP client can operate websites with browser state returned to the agent.

Browser AutomationMCP
Browser Automation MCP Security Reviewed
⭐ 827 GitHub stars
COPY SKILL INSTRUCTIONS (OPTIONAL)
npx skills add agentskillexchange/skills --skill browse-from-mcp-clients-with-browser-use-mcp-server Copy
Uses the third-party skills CLI, not an ASE-owned installer. Check your agent’s compatibility. This copies instructions; complete the upstream tool setup below separately.
View source
At a glance
Tools required
Python, uv, Playwright, mcp-proxy, MCP-compatible client such as Cursor
Install & setup
Install uv, install mcp-proxy with uv tool install mcp-proxy, install the browser-use-mcp-server package, configure the required environment variables, then add the server command to the MCP client configuration.
Author
kontext-security
Publisher
Organization
Last updated
Jul 7, 2026
Quick brief

Use Browser Use MCP Server when an MCP-capable agent needs a browser-use-backed web automation loop rather than a normal manual browser session or direct Playwright scripting. The operator workflow is concrete: install the server and browser prerequisites, configure the MCP client, send browsing tasks from Cursor or another client, and inspect the returned browser state and task output. The scope boundary is the MCP bridge for browser-use driven website operation, not a generic listing for browser-use, Playwright, or a hosted browser product.

How it works

What this skill actually does

Inputs and prerequisites: Python, uv, Playwright, mcp-proxy, MCP-compatible client such as Cursor.

Setup notes: Install uv, install mcp-proxy with uv tool install mcp-proxy , install the browser-use-mcp-server package, configure the required environment variables, then add the server command to the MCP client configuration.

Source and verification boundary: use https://github.com/kontext-security/browser-use-mcp-server as the canonical reference before running the workflow; keep commands, API calls, CLI usage, and generated outputs reviewable against that upstream source.

Framework fit: publish this as a MCP workflow only when the operator can invoke the documented toolchain directly, rather than treating the upstream project as a generic product listing.