The Agent Skills Directory

[SAFE]: The skill consists of instructional text and configuration files that define a software testing process. It contains no executable scripts, binaries, or automation code.\n- [SAFE]: No patterns of data exfiltration, credential access, or network communication were found. The skill operates purely within the context of providing guidance to the agent.\n- [SAFE]: The configuration in SKILL.md and agents/openai.yaml includes restrictive safety flags, such as disabling model invocation and requiring explicit invocation, which follow the principle of least privilege.

regression-test