PISIGuard: Protect your personal and sensitive info when you chat with AI
Summary
PISIGuard is an open-source browser extension that detects and masks personal or sensitive information locally before it is sent to AI chat platforms, then restores the real values in the AI's reply.
View Cached Full Text
Cached at: 08/03/26, 10:31 AM
mohamed–abdel-maksoud/pisiguard
Source: https://github.com/mohamed–abdel-maksoud/pisiguard
PISIGuard
Protect your personal and sensitive info when you chat with AI
PISIGuard: Personally Identifiable & Sensitive Information Guard
I built PISIGuard because I’ve been horrified at how much personal & sensitive information people (including myself) are feeding AI on daily basis. It is cumbersome to censor information by hand, so obviously very few do that at varying degrees of consistency. I wanted something practical that will hide personal & sensitive data, and of course without routing them through yet another server. So PISIGuard runs entirely in your browser. It spots names, email addresses, phone numbers, credit card numbers, passwords, API keys, and more, replaces them with safe placeholders, then puts the real values back into the AI’s reply. From your point of view, it’s mostly invisible; from the AI’s point of view, the sensitive stuff was never there.
All detection, masking, and restoration happens locally. No data is sent to an external service. Nobody sees your messages but you and the AI, and the AI only sees the placeholders.
DISCLAIMER: This software is provided “AS IS”, without warranty of any kind, express or implied, including but not limited to the warranties of merchantability, fitness for a particular purpose and noninfringement. In no event shall the authors or copyright holders be liable for any claim, damages or other liability, whether in an action of contract, tort or otherwise, arising from, out of or in connection with the software or the use or other dealings in the software.
Features
- Catches common sensitive information before they hit the AI servers: names, emails, phone numbers, passwords, API keys, and more.
- Masks each piece of sensitive data with a unique placeholder (like
John Smithinstead of your real name) before the message is sent. - Restores the original values in the AI’s response automatically, so you can read the reply naturally.
- Works on popular AI chat platforms (ChatGPT, Claude, and DeepSeek).
- All processing stays on your device. No telemetry, no server calls, no analytics.
- Minimal permissions: the extension only needs access to the sites you use it on.
- No background processes: the extension only activates when you’re on a supported AI chat page.
- PISIGuard is Configurable: if you need more detailed rules to detect sensitive information, you can provide your own set (advanced use).
Install
From Web Store
From Github Releases
- Download the latest release
.zipfor your browser family (Chrome includes Edge, Opera, etc.; Mozilla includes Firefox). - Unzip the file.
- Load the extension manually:
- Chrome / Edge / Brave
Go to
chrome://extensions, enable “Developer mode”, click “Load unpacked”, and select the unzipped folder. - Firefox
Go to
about:debugging, click “This Firefox”, then “Load Temporary Add-on”, and pick any file inside the unzipped folder.
- Chrome / Edge / Brave
Go to
Build Your Own
- Clone this repo.
- Install the dependencies:
pnpm install - Run the tests:
pnpm test - Build the extension:
pnpm run buildAdistfolder will appear with subfolders for each browser. - Load the unpacked extension from the
distfolder using the same manual steps as above.
PISIGuard is a work of :heart: by Codoma.tech.
Similar Articles
How to Use AI With Your Privacy Intact
The article discusses the privacy risks of AI chatbots like ChatGPT, Claude, and Gemini, and highlights tools such as Confer that use encryption to protect user conversations, providing a guide on using AI while safeguarding personal data.
@GPTdefender: AI prompts can contain more personal data than you think. We catch it before you hit send.
GPT Defender is a browser tool that automatically scans ChatGPT prompts in real time for personal and sensitive data, alerting users before any information is sent. It offers options to review, redact, or allow detected sensitive content before submission.
A way to escape AI surveillance
A guide to disabling AI training data collection across popular chatbots like ChatGPT, Gemini, Claude, and Copilot, along with a mention of a custom extension called UnSee to simplify the process.
@LinusEkenstam: Stop giving your real name, address and other sensitive data to your AI @agentcloakai uses Rampart, the 14.7MB model fr…
AgentCloak is a free browser extension that uses a small AI model to locally detect and replace personal details in AI prompts, ensuring privacy across services like ChatGPT and Claude.
How ChatGPT learns about the world while protecting privacy
OpenAI explains how ChatGPT learns from public data and user interactions while protecting privacy through filtering and user controls.
