Show HN: Local Privacy Firewall-blocks PII and secrets before ChatGPT sees them (github.com)

PrivacyFirewall has launched as an innovative local-first tool designed to protect sensitive information by blocking personally identifiable information (PII) and other secrets before they are sent to AI platforms like ChatGPT, Claude, and Gemini. The tool offers two modes: a lightweight "Lite Mode," which employs regular expressions for quick detection, and an "AI Mode" that utilizes a local Transformer model for deeper PII identification. Crucially, all data processing occurs on the user's device, ensuring no information leaves the local environment, making it a significant advancement in privacy protection for AI interactions. This initiative is particularly relevant as modern AI tools often facilitate unintended data leaks, including emails, API keys, and other sensitive information. Traditional data loss prevention (DLP) systems typically do not cover AI chat prompts, positioning PrivacyFirewall as a robust solution in the current landscape where privacy and data security are paramount. With features such as regex checks, contextual warnings, and auditable open-source software, PrivacyFirewall enhances user security by providing a proactive, zero-trust framework that intercepts sensitive data in real time. The tool not only empowers users to maintain control over their data but also addresses a critical gap in the AI/ML sector concerning privacy and data leakage.