π AI-Safe: Secret Key Detection & Redaction
AI-Safe now supports automatic detection and redaction of secret keys in chat to enhance data security and prevent leakage of sensitive credentials.
π Whatβs New?β
AI-Safe can now identify and redact a wide variety of secret keys across supported file types. This update ensures that your chats are scrubbed of credentials before sharing or further processing.
π Detected Secret Keysβ
The following types of secrets are automatically detected and securely redacted from chats:
| π Secret Type | β Detected |
|---|---|
| JSON Web Token (JWT) | β |
| Google API Key | β |
| OpenAI API Key | β |
| GitHub Personal Access Token (PAT) | β |
| AWS Access Key ID | β |
| Google Client Secret | β |
| Stripe Secret Key | β |
| Stripe Public Key | β |
| Razorpay Key | β |
| Square App ID | β |
| Slack Token | β |
| Telegram Token | β |
| Twilio Messaging SID | β |
| Google Analytics Measurement ID | β |
| Google Analytics Tracking ID | β |
| HuggingFace Token | β |
All of these are automatically redacted during file anonymization runs.