Skip to content

Security: moralfish/maatml

SECURITY.md

Security Policy

Supported Versions

Security fixes are applied to the latest 0.x release of maatml.

Version Supported
0.x Yes

Trust model

maatml treats a model folder as trusted, executable code. plugins: entries in model.yml are imported when the folder is loaded, so every subcommand that reads model.yml, including maatml validate, maatml plan, maatml prepare, maatml train, maatml evaluate, maatml export, and maatml serve, can execute code shipped in that folder. Do not run maatml against model folders from untrusted sources. There is no sandbox; a model folder has the same power as any local Python package you choose to import. maatml validate --no-plugins checks the schema and paths without importing plugin code.

Reporting a Vulnerability

Please do not open a public GitHub issue for security reports.

Prefer one of:

  1. GitHub private vulnerability reporting (preferred): use the Security Advisories flow on this repository when available.
  2. Email: nedal.elghamry@gmail.com

Include enough detail to reproduce the issue (affected version or commit, environment, and steps). You can expect an acknowledgement when the report is received; remediation timelines depend on severity and scope.

There aren't any published security advisories