CookieGuard for Browser Use

A successful agent login is not a secure session cookie. CookieGuard for Browser Use watches the browser your Browser Use agent drives, checks the authentication cookies it receives, and returns PASS / FAIL / INCOMPLETE with rule IDs, explanations, and fixes. The decision comes from the local CookieGuard policy engine, never from the model.

Overview · Install · Policy · How it works · Results · Limits & privacy · Demo & tests

Your policy decides what fails

fail_on sets the line: a confirmed finding at or above it fails the run; anything below is reported. Only cookies you name in auth_cookies can be confirmed. All options and a stricter example: Policy.

See it run

Claude (claude-opus-5-5) drives a real Browser Use agent through the same login twice. The insecure app's cookie lacks HttpOnly, so CookieGuard returns FAIL — CG001 (exit 1). The secure app gets PASS (exit 0). Recorded from a real run: Demo & tests.

Add it to your agent

from browser_use import Agent, BrowserProfile, BrowserSession, Tools
from cookieguard_browser_use import CookieGuardAudit

audit = CookieGuardAudit(
    allowed_origins=['http://127.0.0.1:3000'],  # exact origins you are authorized to test
    auth_cookies=['session_id'],                # the real session cookie name(s)
    fail_on='high',
)
session = BrowserSession(browser_profile=BrowserProfile(
    keep_alive=True,                            # required: Agent.run() otherwise closes the browser
    enable_default_extensions=False,            # avoid the bundled cookie-consent extension
    allowed_domains=audit.browser_allowed_domains(),
))
tools = Tools()
audit.register_tools(tools)                     # optional agent-visible check

await session.start()
await audit.start(session)                      # before the login
agent = Agent(task=..., llm=..., browser_session=session, tools=tools,
              sensitive_data={'http://127.0.0.1': {'login_password': '...'}})
await agent.run()
result = await audit.finish()                   # decided by the CookieGuard engine
await session.kill()

print(result.summary)
result.assert_policy()                          # raises CookiePolicyError on FAIL or INCOMPLETE

assert_policy() raises even if the agent never called the tool, or reported success. CookieGuardAudit is also an async context manager; close() stops collection without a result. One instance audits one workflow.

Needs Python 3.11+, Node.js 24+, and Chrome or Chromium: Install.

Status: 0.1.0b1 pre-release, tested with browser-use 0.13.14. This is a community integration, not an official Browser Use package.