The Agent Skills Directory

[SAFE]: The skill instructions are limited to a single directive to read the file .agents/workflows/exec-plan.md. This is a standard and safe way to structure agentic workflows within a repository.
[SAFE]: The frontmatter configuration disable-model-invocation: true is a safety restriction that limits the execution environment, which does not present a security risk.
[SAFE]: No network operations, sensitive file access, or command execution patterns were found in the provided content.

exec-plan