Browser-Use + GateSolve

Install

pip install gatesolve

Python SDK

Three lines to solve any CAPTCHA. Works with any Python agent framework.

from gatesolve import GateSolve

solver = GateSolve(api_key="gs_your_key_here")

# Solve any CAPTCHA by URL + siteKey
result = solver.solve(
    type="cloudflare-turnstile",
    page_url="https://example.com/login",
    site_key="0x4AAAAAAA..."
)

print(f"Token: {result.token}")  # inject into your form

With Browser-Use

When your browser-use agent encounters a CAPTCHA, call GateSolve to get a valid token.

from browser_use import Agent, Browser
from gatesolve import GateSolve
from langchain_openai import ChatOpenAI

solver = GateSolve(api_key="gs_your_key_here")

async def solve_captcha_task():
    # When your agent hits a CAPTCHA, call GateSolve
    agent = Agent(
        task="Go to example.com/signup and create an account",
        llm=ChatOpenAI(model="gpt-4o"),
    )

    # If the agent encounters a Turnstile challenge:
    result = solver.solve(
        type="cloudflare-turnstile",
        page_url="https://example.com/signup",
        site_key="0x4AAAAAAA..."  # from page source
    )

    # Inject the token into the page
    # The agent can use this token to bypass the CAPTCHA
    print(f"Solved in {result.solve_time}")

MCP Server (Zero Code)

Add GateSolve as an MCP tool. Any MCP-compatible agent can solve CAPTCHAs without writing integration code.

# Or use GateSolve MCP server — zero code integration
# Add to your MCP config:
{
  "mcpServers": {
    "gatesolve": {
      "command": "npx",
      "args": ["-y", "@gatesolve/mcp-server"],
      "env": {
        "GATESOLVE_API_KEY": "gs_your_key_here"
      }
    }
  }
}

# Your agent can now call solve_captcha as a tool

How It Works