ChatShield - AI Data Leak Protection

Stop sensitive data leaks to ChatGPT, Claude & Gemini. Detects API keys, SSNs, emails before sending. 100% local, zero tracking.

STOP LEAKING SECRETS TO AI CHATBOTS Every day, professionals accidentally paste API keys, passwords, and personal data into ChatGPT, Claude, and Gemini. That sensitive information gets sent to third-party servers where it may be stored, logged, or used for training. ChatShield catches these mistakes before they happen. ----- HOW IT WORKS ChatShield scans your input in real-time as you type. When sensitive data is detected, you see a warning before anything leaves your browser: - CANCEL to edit your message - REDACT to automatically mask sensitive data - PROCEED if you meant to include it All detection happens locally on your device. Your data never touches our servers. ----- WHAT WE DETECT Personal Information: - Email addresses - Phone numbers - Social Security Numbers (SSN) - Credit card numbers (with Luhn validation) - Physical addresses Developer Secrets: - AWS access keys and secret keys - GitHub tokens - Stripe API keys (live and test) - OpenAI API keys - Slack tokens - Google Cloud credentials - And more with Pro... ----- SUPPORTED AI PLATFORMS Works on the AI tools you use every day: - ChatGPT (chat.openai.com and chatgpt.com) - Claude (claude.ai) - Gemini (gemini.google.com) We only request access to these 4 specific sites - not all websites. This minimizes permissions while maximizing your privacy. ----- FREE FEATURES Get started with powerful protection at no cost: - Real-time detection of emails, phone numbers, SSNs, credit cards - Detection of 3 common API key types (AWS, GitHub, OpenAI) - Basic redaction with [REDACTED] replacement - 24-hour detection history - Works on ChatGPT, Claude, and Gemini - 100% local processing - your data never leaves your browser - No account required ----- PRO FEATURES ($6/month or $49/year) Unlock maximum protection for power users: Everything in Free, plus: - Smart Fake Data Substitution - Replaces real data with realistic fakes to maintain AI context (e.g., jane.doe@example.com instead of your real email) - ALL API Key Types - Expanded detection for Stripe, Slack, Google, Azure, and more - Custom Regex Patterns - Add your own detection rules for proprietary data formats - Allowlist - Mark specific strings as safe to skip false positives - Unlimited Detection History - Full audit trail of everything blocked - Export Reports - CSV/JSON exports for compliance documentation - Priority Support Start with a FREE 7-day Pro trial - no credit card required. ----- TEAM PLANS ($12/user/month) For organizations that need centralized control: Everything in Pro, plus: - Admin Dashboard - Monitor team-wide detection activity - Usage Analytics - Understand data leak risks across your organization - Centralized Policy Management - Set detection rules for all users - Priority Support with SLA Contact us at team@chatshield.app for team plans. ----- WHY CHATSHIELD? PRIVACY BY DESIGN Unlike cloud-based security tools, ChatShield processes everything locally in your browser. We never see your data. We never phone home. We collect zero analytics. Your sensitive information stays on your device. MINIMAL PERMISSIONS We only access 4 specific AI chat sites - not <all_urls> like many extensions. Check our manifest yourself: we request only "storage" permission for settings and specific host permissions for the AI platforms. DEVELOPER-FIRST Built by developers who understand the risk of accidentally pasting an API key into ChatGPT. Detection patterns are optimized for real-world developer secrets, not just consumer PII. LIGHTWEIGHT No performance impact. Runs efficiently in the background without slowing down your AI conversations. ----- USE CASES DEVELOPERS Stop before pasting that .env file or config with production credentials. ChatShield catches AWS keys, database passwords, and API tokens before they reach AI training data. HEALTHCARE PROFESSIONALS Protect patient information when using AI for documentation or research. Detect PHI before it leaves your organization. LEGAL AND FINANCIAL Keep client data confidential. Catch SSNs, account numbers, and privileged information before sending to third-party AI services. REMOTE WORKERS Prevent accidental disclosure of internal company information, project codenames, or confidential communications. ----- FREQUENTLY ASKED QUESTIONS Q: Does ChatShield send my data anywhere? A: No. All detection happens locally in your browser using JavaScript. Your data never leaves your device. We have no servers to receive your data even if we wanted to. Q: Why do you need host permissions for AI sites? A: We inject a content script to scan the input field before submission. This requires permission to access those specific pages. We intentionally limit this to only 4 AI chat sites. Q: What if I want to send sensitive data intentionally? A: Click "Proceed Anyway" on the warning modal. ChatShield warns you but never blocks you from making your own decisions. Q: Is the source code available? A: We believe in transparency. Our detection logic is inspectable in the extension package itself. ----- PRICING COMPARISON Prevent one accidental API key leak: PRICELESS ChatShield Pro for a year: $49 That leaked AWS key could cost thousands in unauthorized charges. A single data breach can destroy customer trust. ChatShield pays for itself the first time it catches something. ----- GETTING STARTED 1. Install ChatShield (free) 2. Visit ChatGPT, Claude, or Gemini 3. Start typing - ChatShield monitors automatically 4. See warnings when sensitive data is detected 5. Upgrade to Pro for maximum protection No configuration needed. Works instantly after installation. ----- SUPPORT Documentation: https://chatshield.app/help Email: support@chatshield.app Feature requests: https://chatshield.app/feedback ----- Install ChatShield now. Stop leaking secrets to AI chatbots.