Skip to content

Security: ThatLabs2049/that-gpt

Security

SECURITY.md

Security Policy

Supported versions

Version Supported
2.5.x Yes
< 2.5 No

Security fixes are applied to the latest release on the main branch.

Reporting a vulnerability

Please do not open public GitHub issues for security vulnerabilities.

Instead:

  1. Email the maintainers at mohammad161186@gmail.com — or use GitHub Private Vulnerability Reporting if enabled.
  2. Include:
    • Description of the issue and impact
    • Steps to reproduce
    • Affected version(s)
    • Proof of concept if available

We aim to acknowledge reports within 72 hours and provide a fix timeline when possible.

Threat model

ThatGPT is a local-first single-user desktop app:

  • No built-in authentication or multi-user access control
  • API keys are stored in a local .env file in the app data directory
  • Chat history is stored as JSON on disk
  • The app proxies requests to user-configured OpenAI-compatible APIs over HTTPS

In scope

  • Credential exposure via IPC, logs, or UI
  • Path traversal or unsafe file operations in Rust repositories
  • Injection via unsanitized inputs sent to the model API
  • Insecure defaults for TLS or provider requests
  • Supply-chain issues in release artifacts (tampered binaries)

Out of scope

  • Compromise of the user's machine outside the app
  • Attacks requiring physical access to an unlocked PC
  • Misconfiguration of third-party API keys or provider endpoints
  • Social engineering

User recommendations

  • Keep AI_API_KEY only in %APPDATA%\com.thatgpt.desktop\.env — never in the repo or chat exports
  • Verify release downloads using docs/TRUST.md
  • Prefer building from source for maximum assurance
  • Do not run unsigned binaries from untrusted mirrors

Disclosure policy

When a fix is available:

  1. Patch is released and tagged
  2. Advisory is published with credit to the reporter (unless anonymity is requested)
  3. SHA256.txt is regenerated for the new release

Thank you for helping keep ThatGPT secure.

There aren't any published security advisories