The Agent Skills Directory

[PROMPT_INJECTION]: The skill possesses an indirect prompt injection attack surface as it is designed to ingest and analyze untrusted content (source code and git diffs) while retaining the ability to orchestrate further agent actions.
Ingestion points: Reads local source files and git diff outputs provided during runtime as directed by the instructions in SKILL.md.
Boundary markers: The instructions do not define specific delimiters or "ignore instructions" warnings to wrap the code being reviewed.
Capability inventory: Uses the Task tool with subagent_type: general-purpose to perform specialized reviews.
Sanitization: No automated sanitization or filtering of the code content is specified before processing.
[EXTERNAL_DOWNLOADS]: References documentation and design philosophies from well-known sources, including Kent C. Dodds' personal technical site and the official Testing Library website.

reviewing-code