Skip to content

Security: RandomCatUser/Tiny-Me

Security

SECURITY.md

Security Policy

Project: Tiny Me purrGpt

Tiny Me is designed to be a local, privacy-friendly AI chatbot.
We take security and user privacy seriously — please review the following guidelines before reporting issues or vulnerabilities.


Reporting a Vulnerability

If you discover a security issue, please do not open a public issue.

Instead, contact us directly at:

📧 randomcatuser@proton.me

Provide a clear description of the problem, including:

  • Steps to reproduce the issue
  • Possible impact or data affected
  • Suggested fixes or mitigation (if any)

We will acknowledge your report within 48 hours and provide a fix or response within 7–10 days.


Supported Versions

We actively maintain the most recent main branch only.

Version Supported Notes
main (latest) ✅ Yes Actively maintained
older versions ❌ No Use the latest for security patches

Data & Privacy

  • No personal data is sent to any third-party servers except Pollinations API for AI text generation.
  • Local memory (chat history) is stored only in your browser’s localStorage within the PyQt environment.
  • Clearing memory removes all stored data permanently.
  • The app does not collect analytics, telemetry, or external logs.

Safe Usage Tips

  • Avoid entering private, financial, or sensitive data into the chatbot.
  • Always verify API endpoints and source code before connecting to third-party services.
  • Keep your Python environment and dependencies up to date.

Responsible Disclosure

We encourage ethical security research and responsible disclosure.
If your finding results in a confirmed vulnerability, we’ll credit you in the project’s contributors list (if you wish).


Future Security Enhancements

  • Optional encryption for localStorage memory
  • User-level permissions and sandboxing

Thank you for helping us keep Tiny Me purrGpt safe, private, and trustworthy.

There aren’t any published security advisories