NativeMind: An Open-Source Local AI Assistant Browser Extension 131

NativeMind is an open-source localized AI assistant browser extension that supports two local models, Ollama and WebLLM. All data processing is completed locally, effectively protecting user privacy.

Core Features:

• Fully Localized: Data processing and AI computations are performed entirely on local devices, with no need to upload data to the cloud, ensuring privacy and security.

• Open Source: The code is open source, allowing users to freely view and audit it, ensuring transparency.

• High Performance: Supports multiple open-source models, such as DeepSeek, Qwen, Llama, Gemma, and Mistral.

• Seamless Integration: Integrates with Ollama, allowing users to quickly load and switch between different models.

Functions:

• Web Page Summary: Quickly summarizes web page content to help users understand long articles.

• Document Analysis: Supports uploading PDF and Word documents to extract key clauses.

• Bilingual Translation: Provides translation functionality for selected text or entire pages.

• Writing Assistance: Provides writing assistance features such as email polishing and report optimization.

• Multi-Tab Context: Supports simultaneous analysis of multiple web pages while maintaining contextual coherence.

• Floating Toolbar: Call AI features anytime on any page.

• Keyboard Shortcut Support: For example, Ctrl+Shift+L to quickly translate selected text.

• Automatic History Saving: Stores conversation records locally, with optional encryption.

Technical Architecture:

• Dual-Mode Operation Mechanism: Supports Ollama integration (for high-performance needs) and WebLLM support (for quick experience).

Installation Guide:

• Preparations: Requires Chrome browser (latest version), a device with 8GB+ RAM, and at least 5GB of disk space.

• Installation Process:

• Clone the project using Git: git clone https://github.com/NativeMindBrowser/NativeMindExtension.git.

• Enter the project directory: cd NativeMindExtension.

• Install dependencies: pnpm install.

• Start development mode: pnpm dev.

• Installation for Regular Users:

• Visit the Chrome Web Store, find "NativeMind", and click "Add to Chrome".

Privacy Protection:

• Data Closed Loop: Input, processing, and output are all completed in memory, with no network transmission.

• Zero Network Transmission: All external API calls are disabled.

• No User Tracking: Does not collect usage behavior data.

• Local Storage Encryption: History records are encrypted using AES-256.

Usage Steps:

1. Install the Extension: Open the Chrome Web Store, search for "NativeMind", and click "Install".

2. Configure the Model: For first-time use, you need to configure a local model, such as Ollama or WebLLM. Specify the model path or make relevant settings according to the extension's prompts.

3. Use the Smart Conversation Assistant: Click the extension icon, enter questions or commands, and it supports multi-tab contextual understanding.

4. Analyze Web Page Content: While browsing a web page, click the extension icon and select the web page content analysis feature to quickly obtain key information.

5. Full-Page Translation: Click the extension icon and select the full-page translation feature. You can set the target language to provide bilingual comparison.

6. AI Search Enhancement: When searching in the browser, the extension directly provides intelligent search results.

7. Writing Assistance: On pages where text is being edited, the extension automatically detects the text and provides rewriting suggestions.

8. Use the Floating Toolbar: Quickly call AI features on any website without frequently clicking the extension icon.

Open-Source Address:

https://github.com/NativeMindBrowser/NativeMindExtension