The Agent Skills Directory

[SAFE]: The skill performs legitimate quality assurance tasks using standard shell utilities. All operations are confined to the local filesystem and do not involve network exfiltration or unauthorized credential access.
[COMMAND_EXECUTION]: The skill uses shell commands like grep, wc, and test to perform validation. These commands are used for structural analysis (e.g., counting lines, checking for file existence) within the local workspace and represent standard developer workflow behavior.
[DATA_EXPOSURE]: While the skill reads files from the project directory, it does so to generate a local quality report. There is no evidence of sensitive data harvesting or transmission to external domains.
[INDIRECT_PROMPT_INJECTION]: The skill analyzes external content (the skill being checked). It mitigates risks associated with processing untrusted data by using descriptive analysis tools rather than interpreting or executing the content of the files it validates.

quality-checker