Prompt Leak Guard 作者: Dankerbadge Tools
Local-only prompt scanner that catches secrets, tokens, and private data before you send them to AI chats.
擴充套件後設資料
畫面擷圖
關於此擴充套件
Prompt Leak Guard helps AI users and developers check prompts, logs, and config snippets before sending them to AI chat tools.
It includes a popup scanner for pasted text and a supported-site warning script for ChatGPT, Claude, Gemini, and Perplexity. It checks for common API keys, service tokens, private key blocks, webhooks, signed URLs, credential-bearing database URLs, payment-card patterns, emails, phones, and dashed US SSNs. It can also generate sanitized text locally.
Privacy model: scanning runs locally in the browser. There is no backend scanner, analytics SDK, remote model call, or prompt upload.
Important limit: Prompt Leak Guard is a pattern-based safety check, not formal DLP. It cannot guarantee detection of every possible secret or private-data format. If a real key, token, password, or private key may have been exposed, rotate it immediately.
It includes a popup scanner for pasted text and a supported-site warning script for ChatGPT, Claude, Gemini, and Perplexity. It checks for common API keys, service tokens, private key blocks, webhooks, signed URLs, credential-bearing database URLs, payment-card patterns, emails, phones, and dashed US SSNs. It can also generate sanitized text locally.
Privacy model: scanning runs locally in the browser. There is no backend scanner, analytics SDK, remote model call, or prompt upload.
Important limit: Prompt Leak Guard is a pattern-based safety check, not formal DLP. It cannot guarantee detection of every possible secret or private-data format. If a real key, token, password, or private key may have been exposed, rotate it immediately.
由 1 位評論者給出 0 分
權限與資料
必要權限：
- 存取您在 chat.openai.com 的資料
- 存取您在 chatgpt.com 的資料
- 存取您在 claude.ai 的資料
- 存取您在 gemini.google.com 的資料
- 存取您在 www.perplexity.ai 的資料
收集下列資料：
- 開發者聲稱此擴充套件不要求收集任何資料。
更多資訊