Nanobrowser is a free, open-source Chrome/Edge extension that uses multiple collaborating LLM-based agents to automate web tasks through natural language. Users can connect their own LLM providers, including local models via Ollama.
Project overview
It provides a no-code alternative to subscription-based browser automation tools by allowing users to bring their own API keys or run local models for web task execution.
Project type
AI Agent · Workflow
Use cases
Automation
Deployment
Refer to project documentation
License
Apache-2.0
Best for
General users and developers seeking a browser extension for automating repetitive tasks in Chrome or Edge who want to use their own external or local LLM API keys.
Key capabilities
Uses specialized AI agents that collaborate to complete web workflows.
Provides a chat interface with real-time status updates during task execution.
Allows users to access and manage agent interaction history.
Enables users to connect preferred LLM providers and assign different models to different agents.
Automates repetitive web tasks across websites using natural language commands.
Allows running models locally via Ollama or custom OpenAI-compatible providers.
Limitations and risks
Firefox, Safari, and other Chromium variants such as Opera and Arc are not supported.
Local models require more specific and cleaner prompts; ambiguous commands should be avoided.
Third-party derivative projects involving cryptocurrencies, tokens, or NFTs based on the codebase are unaffiliated; users interact with them at their own risk.
Using cost-effective local model configurations may produce less stable outputs and require more task iterations.
Getting started
Install the extension from the Chrome Web Store by clicking Add to Chrome, then configure the Agent Models by adding LLM API keys or local model endpoints.
Load the unpacked extension folder manually into the browser without coding skills, then provide the required LLM model configuration for the Planner and Navigator agents.
Alternatives and comparisons
Enables an AI agent to use a web browser like a human through natural language instructions via a Python library.
An open-source multimodal AI agent stack combining GUI agent, browser automation, and desktop control.
Converts websites and browser sessions into deterministic CLI interfaces for AI agents.