gemini-browser-agent and gemini-computer-use

These are ecosystem siblings—one provides a Chrome extension interface for interactive browser control while the other offers a programmatic Playwright-based automation approach, serving different use cases (manual vs. scriptable) for the same underlying Gemini 2.5 Computer Use capability.

gemini-browser-agent
37
Emerging
gemini-computer-use
29
Experimental
Maintenance 6/25
Adoption 8/25
Maturity 5/25
Community 18/25
Maintenance 6/25
Adoption 6/25
Maturity 1/25
Community 16/25
Stars: 63
Forks: 16
Downloads:
Commits (30d): 0
Language: JavaScript
License:
Stars: 23
Forks: 6
Downloads:
Commits (30d): 0
Language: Python
License:
No License No Package No Dependents
No License No Package No Dependents

About gemini-browser-agent

pmbstyle/gemini-browser-agent

A browser agent with a Google Chrome extension that can work in your browser. Based on Google Gemini 2.5 computer use model.

Bridges a Chrome extension with Google's Gemini Computer Use API to observe and interact with the active tab in real-time, exchanging screenshots and DOM events without requiring sandboxing. Uses a Python WebSocket server that communicates bidirectionally with the extension, enabling the model to execute browser automation tasks directly within your own browser context. Supports agentic workflows where Gemini plans multi-step interactions and streams execution logs back to the UI.

About gemini-computer-use

pmbstyle/gemini-computer-use

A minimal browser automation agent using Google's Gemini 2.5 Computer Use Preview model and Playwright for web browser control.

Implements vision-based browser automation by feeding screenshots to Gemini 2.5 for visual understanding, enabling the model to locate and interact with page elements without DOM parsing. Includes built-in safety guardrails with human-in-the-loop confirmation for sensitive operations, and provides a comprehensive action API covering clicks, typing, scrolling, drag-and-drop, and navigation primitives executed through Playwright's browser control layer.

Scores updated daily from GitHub, PyPI, and npm data. How scores work