🛡️ Transform complex privacy policies into instant insights using Chrome's built-in AI (Gemini Nano).
Automatically detect sign-ups, analyze privacy terms in real-time, and make informed privacy decisions - all LLM insights processed locally in your browser.
Watch our 3-minute demo showcasing how Guardian Sight simplifies privacy policies using Chrome's built-in AI
Key Features:
- ⚡ Transform 10-minute reads into 30-second insights for quick, informed decisions.
- 🚀 Real-time privacy policy detection and analysis
- 🎯 Color-coded privacy risk assessments
- 🔒 Local processing with Chrome's built-in AI
- 🎨 Clear, categorized privacy breakdowns
Before you begin, make sure you have the following installed:
- Node.js (recommended version 16.x or higher) -> Refer
.nvmrcfile to know current version being used. As of now,node: 18.20.4being used. - npm (comes with Node.js)
Follow these steps to get the boilerplate code up and running:
-
Clone the repository:
git clone https://github.com/vivekVells/guardian-sight
-
Navigate to the project folder:
cd guardian-sight -
Install dependencies:
npm install
-
Run the development server:
npm run dev
This command will start the development server using Vite, enabling hot module replacement and allowing you to see your changes in real.
-
Load the extension in Chrome:
- Open the Chrome browser.
- Navigate to
chrome://extensions/. - Turn on the "Developer mode" toggle in the top-right corner.
- Click the "Load unpacked" button and select the
distfolder inside your project directory.
-
Start Developing:
- The popup UI can be found in the
src/App.tsxdirectory. - Customize the manifest template in
src/manifest.jsonand let CRXJS handle the dynamic manifest generation. - The support for background scripts, options page, content scripts is already configured. Customise them in their respective folder.
- The popup UI can be found in the
-
Build for Production:
When you're ready to publish your extension, create a production build by running:
npm run build
The optimized files will be available in the
distfolder. You can then package this folder to distribute your extension.
This diagram illustrates how Guardian Sight transforms privacy policies into actionable insights using Chrome's built-in AI capabilities through multiple LLM roles. Users receive real-time privacy analysis through our intelligent processing pipeline, from initial sign-up detection to final privacy insights presentation.
Key Components:
- LLM as Judge (Prompt API): Detects sign-up flows
- CAG Processor: Chunks and processes privacy policies
- LLM as Generator (Prompt API): Creates initial insights
- LLM as Frontend Helper (Prompt API): Structures data for UI
- LLM as Summarizer (Summarization API): Generates TLDR summaries
Guardian Sight in action across various sign-up flows, demonstrating its versatility and effectiveness in analyzing privacy policies from different services.
Analysis of Google's privacy policy during account creation, highlighting key privacy implications.
Privacy analysis of Claude AI's sign-up flow, showing comprehensive insights into data handling practices.
Breakdown of Facebook's privacy practices during new account registration.
Privacy analysis of Cronitor.io's sign-up process, demonstrating our tool's effectiveness with SaaS platforms.
Each analysis showcases:
- Real-time privacy policy detection
- Comprehensive privacy insights
- Color-coded risk assessments
- Category-specific breakdowns
- TLDR summaries for quick understanding
We welcome contributions to enhance Guardian Sight! Here's how you can get involved:
- Report Issues: Found a bug or have a feature request? Open an issue on our GitHub Issues page.
- Submit Pull Requests: Improve the codebase or add features by creating a pull request. Ensure your contributions align with our contributing guidelines.
Enjoy using Guardian Sight? 🌟 Support our efforts by buying us a coffee! Every contribution helps us keep improving privacy awareness tools.
