Marketplace for trusted AI agent skills

Skill Detail

Control authenticated Chrome sessions through MCP with OpenChrome

Let an agent drive a real logged-in Chrome profile through MCP for authenticated browsing, parallel tab work, and token-efficient page interaction without re-authenticating each run.

Browser AutomationMCP
Browser Automation MCP Security Reviewed
⭐ 206 GitHub stars ⬇ 12.1k/wk npm
COPY SKILL INSTRUCTIONS (OPTIONAL)
npx skills add agentskillexchange/skills --skill control-authenticated-chrome-sessions-through-mcp-with-openchrome 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
Node.js, npm, Google Chrome, MCP-compatible client
Install & setup
Install with npm install -g openchrome-mcp, run openchrome setup for Claude Code or openchrome setup --client codex for Codex CLI, then restart the MCP client. For other clients, configure the MCP server command as openchrome serve --auto-launch.
Author
shaun0927
Publisher
Individual
Last updated
May 22, 2026
Quick brief

Use OpenChrome when an agent needs to operate inside the user’s real Chrome session instead of launching a disposable browser or asking the user to repeat logins. The operator workflow is to install the OpenChrome MCP server, connect it to an MCP-capable client such as Claude Code, Codex, Cursor, VS Code, Windsurf, or OpenCode, and let the agent open isolated tabs, inspect serialized page state, recover from common browser errors, and complete authenticated web tasks against live sites. Invoke this for browser operations where existing cookies, real Chrome behavior, and parallel tab lanes matter more than generic Playwright-style test automation. The scope boundary is the MCP-controlled authenticated browser run: it is not a general browser automation framework listing, and it should be used for explicit agent browser sessions where the user expects the agent to act through Chrome under the configured client.

How it works

What this skill actually does

Inputs and prerequisites: Node.js, npm, Google Chrome, MCP-compatible client.

Setup notes: Install with `npm install -g openchrome-mcp`, run `openchrome setup` for Claude Code or `openchrome setup –client codex` for Codex CLI, then restart the MCP client. For other clients, configure the MCP server command as `openchrome serve –auto-launch`.

Source and verification boundary: use https://github.com/shaun0927/openchrome 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.