Skip to main content
Glama

mcp-helm

Drive your real Chrome session from Claude — copilot mode, with handoff awareness.

Most browser-automation MCP servers spawn a fresh Playwright Chromium. That's useless when the work is "log into Stripe and click 5 things" — you don't have your cookies, your 2FA, your bookmarks. mcp-helm attaches to YOUR Chrome, the one you're already signed into, and lets Claude run a small set of tools against the active tab.

It also knows when to step back: when the page shows a 2FA prompt, captcha, payment confirmation, or biometric request, the screenshot tool flags it and Claude can call handoff() to wait for you.

Why this exists

The eyes-and-hands problem: Claude tells you "click Settings → API access" and you click Settings and there is no API access, so you screenshot back to Claude, which guesses again. That's 5 minutes of round-trips for a 5-second task, and it happens on every Stripe / Apple / Play Console / Cloudflare / Vercel setup.

mcp-helm cuts that loop. Claude sees the actual page, picks elements from the accessibility tree (no coordinate guessing), and stops when it would do something it shouldn't.

Related MCP server: chrome-bridge

Install

npm install -g mcp-helm

Add to ~/.claude.json (or your MCP client's config):

{
  "mcpServers": {
    "helm": {
      "command": "mcp-helm"
    }
  }
}

Usage

1. Launch a driveable Chrome

Add this alias to your shell rc:

alias chrome-pilot='open -a "Google Chrome" --args --remote-debugging-port=9222 --user-data-dir=$HOME/.chrome-pilot'

Run it once: chrome-pilot. A separate Chrome profile opens. Sign into everything you'd want Claude to drive (Play Console, Stripe, etc.). Cookies persist across launches — you only sign in once per service.

Why a separate profile? Your main Chrome can't be launched in remote-debugging mode while it's already running. The dedicated profile lives in ~/.chrome-pilot and stays separate from your daily browsing.

2. From Claude

You: Upload the AAB at <path> to Play Store internal testing.
Claude: [calls helm.attach] → [helm.navigate to play.google.com/console]
        [helm.screenshot] → sees the dashboard
        [helm.click "Personalized AI Portfolio Bot"]
        ... etc

If a 2FA prompt appears, screenshot returns handoffTriggers: ["2FA prompt"] and Claude calls handoff to wait.

Tools

Tool

Purpose

attach

Connect to Chrome on port 9222. Always call first.

list_tabs

List all open tabs.

focus_tab

Switch active tab by index or URL substring.

screenshot

PNG + URL + title + handoff triggers detected.

inspect

Numbered list of interactive elements (a11y tree).

click

Click by id (from inspect), text, or CSS selector. Returns changed: bool from screenshot diff.

type

Type into a field. submit: true presses Enter after.

navigate

Go to a URL.

wait_for

Wait for text or selector.

handoff

Pause and ask the human to take over.

Design choices

  • Accessibility tree, not coordinates. Vision-grounded clicking (Anthropic computer use) is great but flaky on Retina displays and high-DPR scaling. The a11y tree gives stable, semantic IDs — and is what screen readers use.

  • Screenshot diff after every click. If changed: false, the click was a no-op. Saves Claude from cheerfully reporting success.

  • Handoff detection is regex-based, not LLM-based. Cheap, fast, no false positives on common login phrases.

  • No tab-management heuristics. attach picks the first non-blank tab; use list_tabs + focus_tab to be precise. Predictable beats clever.

Status

v0.1 — works for simple flows (Play Console, Stripe dashboard, Vercel, Cloudflare). Edge cases this doesn't handle yet:

  • Shadow DOM components (some web-component-heavy sites)

  • iframes (need to surface frame switching)

  • File uploads from disk

  • Keyboard shortcuts beyond Enter

License

MIT

Available Tools

10 tools
attachA

Connect to a running Chrome instance (must be launched with --remote-debugging-port). Returns the active tab URL and total tab count. Always call this first.

Inputs: port.

clickA

Click by id (from inspect), by visible text, or by CSS selector. Returns {changed: true/false} based on a pre/post screenshot diff — false means the click had no effect.

Inputs: id, text, selector.

focus_tabB

Switch the active tab by index (from list_tabs) or by URL substring.

Inputs: index, urlContains.

handoffA

Pause and ask the human to complete the next step manually (2FA, payment, sensitive config). Returns a paused state — the human resolves it in the browser, then tells Claude to continue.

Inputs: reason (required).

inspectA

Return a numbered list of interactive elements (buttons, links, inputs) from the accessibility tree. Use the returned id with click() or type() — no coordinates needed.

Inputs: no named parameters.

list_tabsA

List all open tabs across contexts. Use to find a specific tab to focus.

Inputs: no named parameters.

screenshotA

Capture the active tab. Returns base64 PNG, URL, title, and any detected handoff triggers (2FA, captcha, payment prompts). Call this after every action to verify state.

Inputs: fullPage.

typeC

Type into a field by id (from inspect) or CSS selector. Set submit=true to press Enter after.

Inputs: text (required), id, selector, submit.

wait_forC

Wait for visible text or a CSS selector to appear on the page.

Inputs: text, selector, timeoutMs.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.6/5.0

    Scored across 10 tools

    Disambiguation5/5

    Each tool has a clearly distinct purpose: attaching, navigating, clicking, typing, inspecting, tab management, screenshotting, waiting, and handoff. No overlapping functionality.

    Naming Consistency5/5

    All tool names follow a consistent verb or verb_noun pattern with lowercase and underscores (e.g., focus_tab, list_tabs, wait_for). No mixing of conventions.

    Tool Count5/5

    With 10 tools, the set covers core browser automation tasks without being bloated or sparse. Each tool earns its place.

    Completeness4/5

    The tool surface covers connection, navigation, interaction, inspection, tab management, and human handoff. Minor gap: no explicit scrolling or back/forward navigation, but these are often handled indirectly.

    Maintenance

    ActivityInactive
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      D
      maintenance
      Transforms Chrome browser into an AI-controlled automation tool that allows AI assistants like Claude to access browser functionality, enabling complex automation, content analysis, and semantic search while preserving your existing browser environment.
      MIT
    • F
      license
      Not graded
      quality
      C
      maintenance
      Provides a direct bridge between Claude and an authenticated Chrome session, enabling fetch requests with session cookies and JavaScript execution within live web pages.
      -
    • A
      license
      Not graded
      quality
      D
      maintenance
      Chrome extension + MCP bridge that gives Claude control over your real browser via CDP, enabling navigation, clicking, typing, scrolling, screenshots, and JS execution with a visible cursor and tab-bring-to-front.
      1
      MIT
    • A
      license
      Not graded
      quality
      B
      maintenance
      Lets Claude Code drive your own Chrome browser with human-like clicks, drags, and typing. It uses your real sessions and logins, and pauses for your confirmation on anything irreversible.
      8
      MIT