Stop sensitive data leaking into ChatGPT, Claude & other AI chats — on-device detection, redaction, smarter prompts.
Users
7
+2 in 30 days
Rating
5.0 / 5
unchanged in 30 days
Ratings
2
0 in 30 days
Store rank
#218K
#182K among extensions
Version
1.0.2
updated 5 Sep 2026
Users
monthly, 2 months
Latest, 17 Oct 2026:7 usersMove the mouse over the chart for any month
Ratings count
monthly, 2 months
Latest, 17 Oct 2026:2 ratingsMove the mouse over the chart for any month
The Chrome Web Store rounds user counts down to the leading digit, so the user line moves in steps. The ratings count is exact and is the better signal of real adoption.
About GuardRail AI – Privacy Shield for AI Chats
GuardRail AI is the invisible safety layer between you and every AI chat you use.
Paste a prompt into ChatGPT, Claude, Gemini, Perplexity, Copilot, or any other AI chat, and GuardRail AI silently scans it before it leaves your browser — catching credit card numbers, SSNs, API keys, private keys, and other secrets before they end up in someone else's training data or logs.
Real protection, not just keyword matching. GuardRail AI validates what it finds instead of guessing, credit card numbers are checked with a real Luhn checksum, API keys are matched against known formats from AWS, GitHub, Stripe, Slack, and OpenAI, and a high-entropy heuristic catches unknown secrets that don't match any known pattern. Add your own custom keywords and regex rules on top for company- or project-specific terms.
On-device, always. Every scan happens locally in your browser. Nothing you type is ever sent to us, to an AI model, or to any third party for scanning — your prompts stay yours. That's not a Pro feature; it's how the whole product works.
Free protection, Pro polish. Detection and blocking are free for everyone — we don't think safety should be behind a paywall. Upgrade to Pro for the productivity layer: redact-and-continue instead of a hard block, prompt chains for multi-step workflows, a command palette (Ctrl/Cmd+K) to insert saved prompts instantly, prompt version history, a drag-and-drop prompt library, and a full analytics dashboard showing your blocked-attempt trends, riskiest sites, and a safety streak.
Your prompt workspace, built in. Save, organize, and reuse your best prompts with variables, categories, and one-click injection directly into the chat box — no more digging through old conversations to find that prompt you wrote last week.
Who it's for: developers who paste code and don't want to leak a key, professionals handling client or company data, and anyone who wants a second set of eyes watching what they send to AI tools.
GuardRail AI — protection first, productivity built in.
https://*.chatgpt.com/*https://*.openai.com/*https://*.claude.ai/*https://*.gemini.google.com/*https://*.perplexity.ai/*https://copilot.microsoft.com/*https://poe.com/*https://chat.mistral.ai/*https://chat.deepseek.com/*https://grok.com/*and 1 more
Alternatives to GuardRail AI – Privacy Shield for AI Chats
Extensions that rank for the same store searches or share the Workflow & Planning category, ordered by users