AI Firewall 作者： Daniel Goldberg

Stop sensitive data from leaking into AI tools - before it leaves your browser.
AI Firewall watches the prompts you submit to AI chat and coding assistants and steps in the moment you're about to send a secret or personal detail. Depending on what it finds, it warns
you, offers a safely-redacted version, or blocks the submission outright - and tells you exactly what it caught and why.
What it detects: API keys and tokens (AWS, GitHub, OpenAI, and more), private keys, credit-card numbers, emails, phone numbers, SSNs, and other personal data.
Supported sites: ChatGPT, Claude, Gemini, Microsoft/GitHub Copilot, Perplexity, Grok, DeepSeek, Mistral, and other major AI services.
Private by design: Inspection happens locally against the AI Firewall proxy on your own machine. Your prompt text is never sent to the developer or any third party. The extension adds no
analytics, tracking, or ads, and ships all of its code in the package (no remote code).
For teams (optional): Organizations can connect the extension to their own self-hosted AI Firewall server to apply company policy and see which AI services are in use. This is off by default and, when enabled, sends only decision metadata - never your prompt text.
Requires the free AI Firewall proxy running locally. Learn more at the homepage link.