The Agent Skills Directory

[COMMAND_EXECUTION]: Uses standard repository utilities (fd, rg, git, gh) to discover and search for documentation files within the project.
[DATA_EXPOSURE]: Accesses repository metadata and file contents, specifically Gherkin feature files and Architecture Decision Records (ADRs), to identify duplicates and inconsistencies.
[PROMPT_INJECTION]: The skill processes untrusted input from file contents which presents an indirect prompt injection surface; however, the impact is minimized by its read-only design and structured reporting format.

review-spec