Skip to content

Security: HDSim-AI/.github

Security

SECURITY.md

Security

Reporting a vulnerability

Email mustafasameen@ufl.edu. Do not open a public issue for a security report.

What counts here

HDSim is a research library. The risks worth reporting are about what it does with your credentials and your data:

  • Credential handling. HDSIM_API_KEY is read from the environment or a .env file. A path where a key is logged, written to disk, or sent anywhere other than the configured HDSIM_BASE_URL is a vulnerability. Report it.
  • Data exfiltration. The library sends persona text to whatever endpoint you configure. A path that sends survey records somewhere you did not configure is a vulnerability.
  • Code execution. Loading a household file or a survey extract must never execute code from it.

What does not count

  • A model returning a wrong or biased decision. That is a research limitation, not a vulnerability; open a normal issue.
  • Cost incurred by running a live simulation against a paid endpoint.

Supported versions

The project is pre-1.0. Fixes land on main; there is no backport branch yet.

There aren't any published security advisories