The Agent Skills Directory

[SAFE]: The skill follows secure engineering practices by implementing a multi-role pipeline with mandatory validation and review steps before completing tasks.
[COMMAND_EXECUTION]: The skill executes project-specific build and test commands (e.g., npm, cargo, go) to verify code integrity. This is a legitimate and necessary function for an architecture optimization tool.
[PROMPT_INJECTION]: While the skill ingests external source code which constitutes an indirect prompt injection attack surface, the risk is mitigated by the structured worker architecture and the requirement for successful validation (e.g., passing builds and tests) of any proposed changes.

team-arch-opt