Skip to content

Security: AleSaiani/agentflow

SECURITY.md

Security Policy

Supported versions

Agent Flow is pre-1.0. Security fixes are applied to the latest released version only.

Version Supported
latest (1.0.0-beta.*)
older pre-releases

Reporting a vulnerability

Please do not open a public issue for security problems.

Report privately through GitHub's private vulnerability reporting for this repository. Include:

  • a description of the issue and its impact,
  • the affected version (or commit),
  • steps to reproduce, and a proof of concept if possible.

You can expect an initial response within a few days. Once a fix is ready, disclosure will be coordinated with you.

Scope notes

Agent Flow runs LLM-driven workflows on your machine and executes shell stages and subagents as instructed by the workflows you author or run. Treat untrusted WORKFLOW.md files as untrusted code: review them before running, exactly as you would a shell script. The engine itself has zero runtime dependencies (Node builtins only), which keeps its dependency attack surface minimal.

There aren't any published security advisories