The Agent Skills Directory

[SAFE]: The skill consists entirely of instructional guidelines for an AI agent acting as a quality assurance reviewer. It defines policies for assessing linting, formatting, and manual verification evidence without including any executable scripts, network operations, or access to sensitive data.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted external data such as code diffs and pull request descriptions. While this represents a theoretical attack surface, the instructions specifically guide the agent to perform a structured, evidence-based review rather than executing arbitrary commands found in the input. No specific exploitable injection vectors were detected in the skill's logic.

quality-assurance