New to Claude Skills? Learn how to install them →

steipete on GitHub

Browser Use

Free

Control your existing Chrome profile for UI testing.

by steipete6.5k stars on steipete/agent-scripts
1 views
Updated Aug 10, 2026
Get this skill

Free · Opens the source repo

What Browser Use does

Browser Use allows developers to interact directly with their signed-in Chrome profile, which is essential for tasks that require authentication or access to user-specific data. This skill is particularly useful for login-dependent work and live UI verification, ensuring that the correct cookies and extensions are utilized during testing. By leveraging the Codex Chrome plugin first, followed by the OpenClaw extension, and falling back to direct DevTools attachment only when necessary, it provides a robust framework for browser automation.

The skill operates through a structured route that prioritizes the most effective method of control, starting with the Codex plugin, which must be callable in the active session. If that is unavailable, it defaults to using the OpenClaw extension, which facilitates a connection to the Chrome DevTools via a secure relay. This ensures that the user's real profile is used, which is crucial for accurately reproducing browser behavior, especially on sites that rely heavily on user authentication and session data.

For developers and testers, this skill streamlines the process of verifying UI changes and interactions in a live environment. It supports a variety of commands through the mcporter interface, allowing users to list pages, select targets, take snapshots, and execute scripts. This level of control is invaluable for debugging and ensuring that changes do not adversely affect the user experience. The focus on maintaining security and privacy by keeping credentials out of logs and outputs further enhances its reliability.

Overall, Browser Use is an essential tool for anyone involved in web development or testing, providing a seamless way to interact with the browser while respecting user privacy and security protocols.

When to use it

Use this skill when you need to perform UI testing or debugging that requires access to a user's signed-in Chrome profile.

When not to use it

Avoid this skill for simple automation tasks that do not require user authentication or when testing in isolated environments is acceptable.

What you can build with it

Testing Login-Dependent Features

Use Browser Use to verify that login-dependent features work correctly by interacting with the live UI using your signed-in profile.

Debugging User-Specific Issues

When a bug is reported that only occurs for certain users, leverage Browser Use to replicate the issue in the actual user environment.

Validating UI Changes

After making changes to the UI, use Browser Use to ensure that the modifications behave as expected in a real browser context.

How to install Browser Use

View source

1. Install with the skills CLI

npx skills add steipete/agent-scripts/browser-use --agent claude-code

2. Or install it manually

Download the skill folder and drop it into ~/.claude/skills/ for all projects, or .claude/skills/ to scope it to one repo. Restart Claude Code so it picks up the new skill.

Anthropic's agentic coding CLI, and the reference implementation of Agent Skills. Drop a skill folder into ~/.claude/skills and Claude Code loads it automatically whenever a task matches the skill's description. Claude Code docs

Inside SKILL.md

Written by steipete

Browser Use

Control the user's existing real Chrome profile, especially for login-dependent work and live UI verification.

Route

  1. Use the Codex Chrome or Chrome [Internal] plugin when it is callable in the active session. Installed on disk is not enough.
  2. Otherwise prefer the OpenClaw extension-backed mcporter route.
  3. Use legacy direct DevTools attachment only as the explicit last fallback.

For mcporter, use only the normal interface:

mcporter call chrome-devtools.<tool>

Never use isolated Chrome, the Codex in-app browser, Playwright, Puppeteer, AppleScript, osascript, generic GUI scripting, or macOS open as a browser-control substitute unless the user explicitly asks for an isolated or new browser. Peekaboo is allowed only for Chrome or extension setup and visible prompts. Login-heavy sites often depend on the real profile's cookies, SSO, device trust, and extensions.

For a rendered-browser bug, prove behavior through this real profile. Treat curl, source inspection, API checks, and isolated test browsers as supporting evidence, not substitutes for live UI proof.

Extension Relay Model

OpenClaw creates a random per-host relay secret in its mode-0600 credentials/ directory. Pairing gives the Chrome extension a relay URL plus that secret. The extension stores both in chrome.storage.local and authenticates its WebSocket through WebSocket subprotocols; the secret is not placed in the request URL.

On a same-host relay, mcporter reads the local secret, probes /json/version with Bearer authentication, then connects chrome-devtools-mcp to the authenticated /cdp endpoint. Keep credentials out of configuration, command output, chat, logs, and screenshots.

The relay exposes only tabs in the Chrome tab group titled OpenClaw. Group membership is the user-visible consent and authorization boundary: putting a tab in shares it; removing the tab revokes access. Group color is irrelevant.

Topology Boundary

Direct remote Gateway pairing over wss:// lets OpenClaw's Gateway-side browser tool control local Chrome. It does not create a local relay for a local mcporter process. Do not routinely copy remote secrets or build ad-hoc SSH tunnels around this boundary.

If local mcporter cannot authenticate to a local relay, the extension-backed mcporter route is unavailable. Report that clearly or use the labeled legacy fallback; never represent remote Gateway control or direct attachment as local relay success.

Setup and Repair

  • Resolve the unpacked extension with openclaw browser extension path, then verify <path>/manifest.json before loading or presenting it. The source fallback is ~/Projects/openclaw/extensions/browser/chrome-extension; verify its manifest too.
  • Pair through the extension popup without printing the pairing string in chat or captured tool output, and never put it in shell history, logs, or screenshots. Paste it only into the confirmed popup and clear any temporary clipboard value.
  • Confirm the popup reports Connected · N tabs shared and that the intended tab is in the OpenClaw group.
  • Restart the mcporter daemon after pairing or changing the relay route, then re-run readiness proof from scratch.
  • Remote pairing with --gateway-url serves the Gateway browser path; it does not make the relay available to local mcporter.

Do not use openclaw browser extension cdp --json or inspect process arguments as routine diagnostics: both can expose relay credentials. If credential exposure is suspected, rotate the per-host secret and pair again.

Fail-Closed Readiness Proof

Require every condition below before calling the extension route ready:

  1. The popup says Connected · N tabs shared.
  2. The intended disposable tab is shared and belongs to the OpenClaw group.
  3. mcporter has been restarted after pairing or relay-route changes.
  4. list_pages matches the shared tab set exactly. If unrelated, unshared tabs appear, mcporter used full-profile legacy attachment; do not count that as extension success.
  5. Navigation and evaluation both succeed in the disposable shared tab.
mcporter call chrome-devtools.list_pages --args '{}' --output text
mcporter call chrome-devtools.select_page --args '{"pageId":9}' --output text
mcporter call chrome-devtools.navigate_page --args '{"url":"https://example.com/?openclaw-relay-proof=1"}' --output text
mcporter call chrome-devtools.evaluate_script --args '{"function":"() => ({title: document.title, href: location.href})"}' --output json

A blocking Allow remote debugging? prompt proves legacy attachment was attempted. Its absence alone does not prove the relay path; the popup, group, exact page set, and read/write checks provide that proof.

Typical Flow

List pages, select only a shared target, snapshot before acting, and use fresh snapshot UIDs. Prefer DOM snapshots over screenshots unless layout matters.

mcporter call chrome-devtools.list_pages --args '{}' --output text
mcporter call chrome-devtools.select_page --args '{"pageId":9}' --output text
mcporter call chrome-devtools.take_snapshot --args '{}' --output text
mcporter call chrome-devtools.click --args '{"uid":"1_38","includeSnapshot":true}' --output text
mcporter call chrome-devtools.fill --args '{"uid":"1_13","value":"text","includeSnapshot":true}' --output text
mcporter call chrome-devtools.evaluate_script --args '{"function":"() => document.title"}' --output json

For live UI proof, capture the current page state before the action, perform the requested interaction, then snapshot or evaluate the rendered result. Keep secrets out of DOM, input, network, console, and screenshot output. For credential checks, return only safe shape such as present/absent, length, status code, or account/organization label.

If automation is unavailable, report the verification gap instead of silently switching to prohibited or isolated tooling.

Argument and Output Mechanics

--args accepts inline JSON only. It does not read @file. Flag-style named arguments do:

mcporter call chrome-devtools.navigate_page url=@/tmp/target-url.txt --output text
mcporter call chrome-devtools.evaluate_script function=@/tmp/probe.js --output json

Treat this as a safety primitive, not just ergonomics. A sign-in URL, magic link, or callback URL is credential-equivalent: write it to a mode-0600 file and pass it by file reference so it never reaches shell history, process arguments, or captured tool output. The same form carries a multi-line JavaScript function without quoting or control-character errors.

Other call mechanics worth knowing before a login flow:

  • The default call timeout is short (about five seconds). Real navigation, snapshots, and consent pages routinely exceed it, and the failure looks identical to a hung page. Pass --timeout 30000 for anything interactive.
  • take_screenshot with filePath is confined to the server's configured workspace roots and refuses arbitrary paths. Omit filePath, read the base64 image from --output json, and decode it locally.
  • new_page can fail with a restricted/unavailable tab when the target is not shared. Navigate an already-shared tab instead of creating one.
  • mcporter list <server> --schema prints the real function signatures. Use it rather than guessing parameter names.

Clicks That Do Not Click

A click by uid can return success and still do nothing: some pages bind their handlers so that a synthetic click is ignored. Silence is not proof of action, so verify state after every activation rather than assuming it worked.

When a click no-ops, drive the control from the keyboard with press_key (Tab, Shift+Tab, Enter) and confirm focus with a screenshot before committing. Focus rings are the only reliable evidence of which control is about to receive Enter, and consent screens routinely put the safe-looking prominent button next to a low-emphasis link that is the one you actually want. Blind Enter on such a page picks the wrong control.

Snapshot uid values are invalidated by any navigation or re-render. Re-run take_snapshot and re-resolve the uid after every step; a stale uid reports that the element no longer exists, which is a cue to re-snapshot rather than to retry the same call.

When the Relay Goes Empty Mid-Task

The relay can stop exposing tabs partway through a task — the shared tab was closed, navigated somewhere the group no longer covers, or the extension dropped its connection. The failure is quiet: list_pages returns an empty result rather than an error, and every later call times out against nothing.

Treat an empty page list as "the relay lost its tabs", not "the browser is gone". Confirm the browser process is actually running before doing anything drastic. Restarting the mcporter daemon does not re-share tabs, because sharing is the user's group membership, not daemon state — so a restart loop cannot fix this and only costs time. Re-establishing access needs the user to put a tab back in the shared group.

Do not escalate to a full-profile attachment or an isolated browser to route around it. When the task is a sign-in the user is present for, the faster and more honest move is to hand the URL to the browser the user is already sitting at, and if the flow's callback listens on another host, forward that port to the user's machine first. Otherwise report the access gap.

Legacy Fallback: Full-Profile Direct Attachment

Use this only after the callable plugin and authenticated local extension relay are unavailable. It exposes the full real-profile tab set and can show Chrome's blocking Allow remote debugging? prompt.

When a visible, unambiguous Chrome prompt asks to allow the attachment, approve it once, then rerun list_pages. If the prompt is absent, ambiguous, or the retry fails, stop and ask the user or report that Chrome DevTools MCP is unavailable. Do not loop approvals, repeatedly restart Chrome or mcporter, or kill browser processes.

Verify that list_pages shows the intended real-profile tabs before acting. Always label this path as full-profile direct attachment, never as OpenClaw extension relay success.

Frequently asked questions about Browser Use

Similar skills