The Agent Skills Directory

[SAFE]: The skill operates as a passive analytical tool for auditing quality standards. It explicitly forbids itself from writing or modifying code, which eliminates the risk of malicious implementation through the agent's actions.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted external data, such as design specifications and feature documentation during its audit phase. However, because the skill lacks executable tools or capabilities (such as shell access, network requests, or file system writes), this ingestion surface does not present a viable attack vector for the agent.
[COMMAND_EXECUTION]: No dangerous commands, shell script patterns, or subprocess execution triggers were detected. The skill's workflow is strictly limited to information synthesis, scorecard generation, and qualitative reporting.

warden