/ README.md
README.md
1 <h1 align="center"> 2 <img src="https://github.com/user-attachments/assets/ec60b0c4-87ba-48f4-981a-c55ed0e8497b" height="100" width="375" alt="banner" /><br> 3 </h1> 4 5 6 <div align="center"> 7 8 [](https://github.com/nanobrowser) 9 [](https://x.com/nanobrowser_ai) 10 [](https://discord.gg/NN3ABHggMK) 11 [<img src="https://deepwiki.com/badge.svg" height="28" alt="Ask DeepWiki">](https://deepwiki.com/nanobrowser/nanobrowser) 12 [](https://github.com/sponsors/alexchenzl) 13 14 </div> 15 16 ## 🌐 Nanobrowser 17 18 Nanobrowser is an open-source AI web automation tool that runs in your browser. A free alternative to OpenAI Operator with flexible LLM options and multi-agent system. 19 20 ⬇️ Get [Nanobrowser from Chrome Web Store](https://chromewebstore.google.com/detail/nanobrowser/imbddededgmcgfhfpcjmijokokekbkal) for free 21 22 👏 Join the community in [Discord](https://discord.gg/NN3ABHggMK) | [X](https://x.com/nanobrowser_ai) 23 24 🌟 Loving Nanobrowser? Give us a star and help spread the word! 25 26 ❤️ Support the project by [sponsoring us](https://github.com/sponsors/alexchenzl) - every contribution helps keep Nanobrowser free and open source! 27 28 <div align="center"> 29 <img src="https://github.com/user-attachments/assets/112c4385-7b03-4b81-a352-4f348093351b" width="600" alt="Nanobrowser Demo GIF" /> 30 <p><em>Nanobrowser's multi-agent system analyzing HuggingFace in real-time, with the Planner intelligently self-correcting when encountering obstacles and dynamically instructing the Navigator to adjust its approach—all running locally in your browser.</em></p> 31 </div> 32 33 ## 🔥Why Nanobrowser? 34 35 Looking for a powerful AI browser agent without the $200/month price tag of OpenAI Operator? **Nanobrowser** , as a chrome extension, delivers premium web automation capabilities while keeping you in complete control: 36 37 - **100% Free** - No subscription fees or hidden costs. Just install and use your own API keys, and you only pay what you use with your own API keys. 38 - **Privacy-Focused** - Everything runs in your local browser. Your credentials stay with you, never shared with any cloud service. 39 - **Flexible LLM Options** - Connect to your preferred LLM providers with the freedom to choose different models for different agents. 40 - **Fully Open Source** - Complete transparency in how your browser is automated. No black boxes or hidden processes. 41 42 > **Note:** We currently support OpenAI, Anthropic, Gemini, Ollama, Groq, Cerebras, Llama and custom OpenAI-Compatible providers, more providers will be supported. 43 44 45 ## 📊 Key Features 46 47 - **Multi-agent System**: Specialized AI agents collaborate to accomplish complex web workflows 48 - **Interactive Side Panel**: Intuitive chat interface with real-time status updates 49 - **Task Automation**: Seamlessly automate repetitive web automation tasks across websites 50 - **Follow-up Questions**: Ask contextual follow-up questions about completed tasks 51 - **Conversation History**: Easily access and manage your AI agent interaction history 52 - **Multiple LLM Support**: Connect your preferred LLM providers and assign different models to different agents 53 54 55 ## 🌐 Browser Support 56 57 **Officially Supported:** 58 - **Chrome** - Full support with all features 59 - **Edge** - Full support with all features 60 61 **Not Supported:** 62 - Firefox, Safari, and other Chromium variants (Opera, Arc, etc.) 63 64 > **Note**: While Nanobrowser may function on other Chromium-based browsers, we recommend using Chrome or Edge for the best experience and guaranteed compatibility. 65 66 67 ## 🚀 Quick Start 68 69 1. **Install from Chrome Web Store** (Stable Version): 70 * Visit the [Nanobrowser Chrome Web Store page](https://chromewebstore.google.com/detail/nanobrowser/imbddededgmcgfhfpcjmijokokekbkal) 71 * Click "Add to Chrome" button 72 * Confirm the installation when prompted 73 74 > **Important Note**: For latest features, install from ["Manually Install Latest Version"](#-manually-install-latest-version) below, as Chrome Web Store version may be delayed due to review process. 75 76 2. **Configure Agent Models**: 77 * Click the Nanobrowser icon in your toolbar to open the sidebar 78 * Click the `Settings` icon (top right) 79 * Add your LLM API keys 80 * Choose which model to use for different agents (Navigator, Planner) 81 82 ## 🔧 Manually Install Latest Version 83 84 To get the most recent version with all the latest features: 85 86 1. **Download** 87 * Download the latest `nanobrowser.zip` file from the official Github [release page](https://github.com/nanobrowser/nanobrowser/releases). 88 89 2. **Install**: 90 * Unzip `nanobrowser.zip`. 91 * Open `chrome://extensions/` in Chrome 92 * Enable `Developer mode` (top right) 93 * Click `Load unpacked` (top left) 94 * Select the unzipped `nanobrowser` folder. 95 96 3. **Configure Agent Models** 97 * Click the Nanobrowser icon in your toolbar to open the sidebar 98 * Click the `Settings` icon (top right). 99 * Add your LLM API keys. 100 * Choose which model to use for different agents (Navigator, Planner) 101 102 4. **Upgrading**: 103 * Download the latest `nanobrowser.zip` file from the release page. 104 * Unzip and replace your existing Nanobrowser files with the new ones. 105 * Go to `chrome://extensions/` in Chrome and click the refresh icon on the Nanobrowser card. 106 107 ## 🛠️ Build from Source 108 109 If you prefer to build Nanobrowser yourself, follow these steps: 110 111 1. **Prerequisites**: 112 * [Node.js](https://nodejs.org/) (v22.12.0 or higher) 113 * [pnpm](https://pnpm.io/installation) (v9.15.1 or higher) 114 115 2. **Clone the Repository**: 116 ```bash 117 git clone https://github.com/nanobrowser/nanobrowser.git 118 cd nanobrowser 119 ``` 120 121 3. **Install Dependencies**: 122 ```bash 123 pnpm install 124 ``` 125 126 4. **Build the Extension**: 127 ```bash 128 pnpm build 129 ``` 130 131 5. **Load the Extension**: 132 * The built extension will be in the `dist` directory 133 * Follow the installation steps from the Manually Install section to load the extension into your browser 134 135 6. **Development Mode** (optional): 136 ```bash 137 pnpm dev 138 ``` 139 140 ## 🤖 Choosing Your Models 141 142 Nanobrowser allows you to configure different LLM models for each agent to balance performance and cost. Here are recommended configurations: 143 144 ### Better Performance 145 - **Planner**: Claude Sonnet 4 146 - Better reasoning and planning capabilities 147 - **Navigator**: Claude Haiku 3.5 148 - Efficient for web navigation tasks 149 - Good balance of performance and cost 150 151 ### Cost-Effective Configuration 152 - **Planner**: Claude Haiku or GPT-4o 153 - Reasonable performance at lower cost 154 - May require more iterations for complex tasks 155 - **Navigator**: Gemini 2.5 Flash or GPT-4o-mini 156 - Lightweight and cost-efficient 157 - Suitable for basic navigation tasks 158 159 ### Local Models 160 - **Setup Options**: 161 - Use Ollama or other custom OpenAI-compatible providers to run models locally 162 - Zero API costs and complete privacy with no data leaving your machine 163 164 - **Recommended Models**: 165 - **Qwen3-30B-A3B-Instruct-2507** 166 - **Falcon3 10B** 167 - **Qwen 2.5 Coder 14B** 168 - **Mistral Small 24B** 169 - [Latest test results from community](https://gist.github.com/maximus2600/75d60bf3df62986e2254d5166e2524cb) 170 - We welcome community experience sharing with other local models in our [Discord](https://discord.gg/NN3ABHggMK) 171 172 - **Prompt Engineering**: 173 - Local models require more specific and cleaner prompts 174 - Avoid high-level, ambiguous commands 175 - Break complex tasks into clear, detailed steps 176 - Provide explicit context and constraints 177 178 > **Note**: The cost-effective configuration may produce less stable outputs and require more iterations for complex tasks. 179 180 > **Tip**: Feel free to experiment with your own model configurations! Found a great combination? Share it with the community in our [Discord](https://discord.gg/NN3ABHggMK) to help others optimize their setup. 181 182 ## 💡 See It In Action 183 184 Here are some powerful tasks you can accomplish with just a sentence: 185 186 1. **News Summary**: 187 > "Go to TechCrunch and extract top 10 headlines from the last 24 hours" 188 189 2. **GitHub Research**: 190 > "Look for the trending Python repositories on GitHub with most stars" 191 192 3. **Shopping Research**: 193 > "Find a portable Bluetooth speaker on Amazon with a water-resistant design, under $50. It should have a minimum battery life of 10 hours" 194 195 ## 🛠️ Roadmap 196 197 We're actively developing Nanobrowser with exciting features on the horizon, welcome to join us! 198 199 Check out our detailed roadmap and upcoming features in our [GitHub Discussions](https://github.com/nanobrowser/nanobrowser/discussions/85). 200 201 ## 🤝 Contributing 202 203 **We need your help to make Nanobrowser even better!** Contributions of all kinds are welcome: 204 205 * **Share Prompts & Use Cases** 206 * Join our [Discord server](https://discord.gg/NN3ABHggMK). 207 * share how you're using Nanobrowser. Help us build a library of useful prompts and real-world use cases. 208 * **Provide Feedback** 209 * Try Nanobrowser and give us feedback on its performance or suggest improvements in our [Discord server](https://discord.gg/NN3ABHggMK). 210 * **Contribute Code** 211 * Check out our [CONTRIBUTING.md](CONTRIBUTING.md) for guidelines on how to contribute code to the project. 212 * Submit pull requests for bug fixes, features, or documentation improvements. 213 214 215 We believe in the power of open source and community collaboration. Join us in building the future of web automation! 216 217 218 ## 🔒 Security 219 220 If you discover a security vulnerability, please **DO NOT** disclose it publicly through issues, pull requests, or discussions. 221 222 Instead, please create a [GitHub Security Advisory](https://github.com/nanobrowser/nanobrowser/security/advisories/new) to report the vulnerability responsibly. This allows us to address the issue before it's publicly disclosed. 223 224 We appreciate your help in keeping Nanobrowser and its users safe! 225 226 ## 💬 Community 227 228 Join our growing community of developers and users: 229 230 - [Discord](https://discord.gg/NN3ABHggMK) - Chat with team and community 231 - [Twitter](https://x.com/nanobrowser_ai) - Follow for updates and announcements 232 - [GitHub Discussions](https://github.com/nanobrowser/nanobrowser/discussions) - Share ideas and ask questions 233 234 ## 👏 Acknowledgments 235 236 Nanobrowser builds on top of other awesome open-source projects: 237 238 - [Browser Use](https://github.com/browser-use/browser-use) 239 - [Puppeteer](https://github.com/EmergenceAI/Agent-E) 240 - [Chrome Extension Boilerplate](https://github.com/Jonghakseo/chrome-extension-boilerplate-react-vite) 241 - [LangChain](https://github.com/langchain-ai/langchainjs) 242 243 Huge thanks to their creators and contributors! 244 245 ## 📄 License 246 247 This project is licensed under the Apache License 2.0 - see the [LICENSE](LICENSE) file for details. 248 249 Made with ❤️ by the Nanobrowser Team. 250 251 Like Nanobrowser? Give us a star 🌟 and join us in [Discord](https://discord.gg/NN3ABHggMK) | [X](https://x.com/nanobrowser_ai) 252 253 ## ⚠️ DISCLAIMER ON DERIVATIVE PROJECTS 254 255 **We explicitly *DO NOT* endorse, support, or participate in any** projects involving cryptocurrencies, tokens, NFTs, or other blockchain-related applications **based on this codebase.** 256 257 **Any such derivative projects are NOT Affiliated with, or maintained by, or in any way connected to the official Nanobrowser project or its core team.** 258 259 **We assume NO LIABILITY for any losses, damages, or issues arising from the use of third-party derivative projects. Users interact with these projects at their own risk.** 260 261 **We reserve the right to publicly distance ourselves from any misuse or misleading use of our name, codebase, or brand.** 262 263 We encourage open-source innovation but urge our community to be discerning and cautious. Please ensure you understand the risks before using any software or service built upon our codebase by independent developers. 264 265