Skip to content

Security: Barbariandev/mantis_model_iteration_tool

Security

SECURITY.md

Security Policy

Reporting Vulnerabilities

If you find a vulnerability, please do not open a public issue with exploit details. Report it privately to the project maintainers through the repository security advisory workflow or the maintainer contact channel listed by the publishing organization.

Include:

  • Affected commit or release.
  • Reproduction steps.
  • Impact and reachable attack surface.
  • Any logs or proof-of-concept code needed to validate the issue.

Operational Expectations

MANTIS can execute agent-generated code and user-submitted strategy code. Run it with the same care you would apply to any system that launches containers and spends API credits.

  • Never expose the GUI publicly without MANTIS_AUTH_TOKEN.
  • Never run the Targon agent server without MANTIS_SERVER_AUTH_KEY.
  • Never run the remote evaluation service without MANTIS_EVAL_API_KEY.
  • Rotate any API key that was stored in a local file before publishing.
  • Keep Docker, Python, Node.js, Claude Code CLI, and deployment images patched.

Secret Handling

The repository ignores local key/config files, caches, agent workspaces, generated deployment files, and wallet artifacts. Before publishing or submitting changes, run:

git status --short --ignored

Confirm that no local secrets or generated runtime data are staged.

There aren't any published security advisories