openspace

SUSPICIOUS: The skill’s core behavior is broadly consistent with its stated purpose, and the install source appears same-org and verifiable. However, it has a larger-than-normal trust footprint for a skill: automatic installation of additional host skills, credential forwarding to external tooling, cloud upload/download of skills, and self-evolving behavior that can modify future agent actions. This is not clearly malicious, but it is medium-to-high risk and should be treated as a powerful external delegation system rather than a simple local helper.