The Agent Skills Directory

[SAFE]: The skill is designed for internal project maintenance, specifically focusing on architectural analysis and refactoring workflows. It operates by reading repository files and writing progress reports to a local plans/ directory.
[SAFE]: Command execution is restricted to standard development tasks, such as running existing project tests and performing git commits. These operations are expected within the context of a software engineering tool.
[SAFE]: The skill does not exhibit any network exfiltration, credential harvesting, or remote code execution patterns. It maintains a strictly local scope and does not attempt to download external dependencies or scripts.

architectural-review