The Agent Skills Directory

[COMMAND_EXECUTION]: The skill mentions the use of a script scripts/skill-audit.js --graded for grading and verification purposes. This script is part of the internal infrastructure of the project as defined in the grounding metadata.
[EXTERNAL_DOWNLOADS]: The skill references documentation and evaluation files located at https://github.com/jacob-balslev/skill-graph. These resources originate from the vendor's own repository and are used for instructional grounding and quality assurance.
[INDIRECT_PROMPT_INJECTION]: As a quality assurance tool, the skill is designed to analyze source code and linting output. While this introduces a theoretical surface for indirect injection via malicious code comments, the skill is primarily instructional and does not automate the execution of untrusted logic without human oversight.

lint-overlay