mksglumksgluCommunity·1 Agent Skills Included

browsirai

Control your real Chrome browser with logged-in sessions

Connects coding agents to your live Chrome browser via the DevTools Protocol to navigate pages, click elements, fill forms, and take screenshots. Reuses your existing logged-in sessions and cookies, eliminating OAuth setup, API tokens, and repeated logins. Returns compact text snapshots instead of heavy screenshots, cutting token costs while keeping credentials safely off the model context.
npx skills add mksglu/browsirai --all -g -y

All Skills in This Repository (1)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install browsirai?

Run `npx skills add mksglu/browsirai --all -g -y` in your terminal to install the skill globally for your agent.

How does browsirai use my logged-in sessions?

It connects to your real Chrome binary over the DevTools Protocol, so your existing cookies and logins for sites like GitHub or Jira work instantly without OAuth or API tokens.

Are my passwords and cookies safe with browsirai?

Yes. Cookie values are copied at the filesystem level and never enter the model context, while auth headers, tokens, and secrets in network data are automatically redacted.

Which browsers and platforms does browsirai support?

It works with Chrome, Edge, Brave, Arc, Vivaldi, Opera, and Chromium on macOS, Linux, and Windows, and integrates with Claude Code, Cursor, VS Code Copilot, Windsurf, and other MCP-compatible clients.

Can I use browsirai without an AI agent?

Yes. It also runs as a standalone terminal CLI with commands like open, snapshot, click, and fill for direct browser control.

Related Repositories in Software Engineering

View All in Software Engineering