Support.

Request support

Email info@fragenta.com and include the PromptGuard version, browser, affected supported AI website, and a short description of the visible error.

Never send real passwords, enrollment codes, device tokens, IBANs, API keys, or complete prompt content. Use clearly artificial test data for examples only.

First safe steps

• Check whether the PromptGuard icon is visible in the browser.
• Use the icon to open settings and verify the protection status.
• If “Register again” appears, request a fresh one-time code from the workspace administrator.
• Reload the AI website once after an extension update.

Report a vulnerability

Report potential vulnerabilities confidentially with the subject “PromptGuard Security” to info@fragenta.com. Please do not publish details while the report is being reviewed and, if necessary, fixed.

Transparency and limitations

The current pilot protects text input on explicitly supported AI websites in managed desktop browsers. It does not protect native apps, direct API calls, images, screenshots, or file uploads.

Details about the data flow are provided in the PromptGuard privacy notice.