The Agent Skills Directory

Indirect Prompt Injection (HIGH): The skill defines a high-severity vulnerability surface because it directs the agent to process untrusted external data (code files) with high-privilege write capabilities.
Ingestion points: External source code and style files accessed via the Read tool specified in the metadata.
Boundary markers: None. The skill does not instruct the agent to use delimiters or ignore instructions found within the processed files.
Capability inventory: Edit and Write tools allow the agent to modify the filesystem based on interpreted content.
Sanitization: None. There is no logic provided to filter or validate the content of the files being read.
No Code (INFO): The skill is purely instructional and contains no executable logic, scripts, or binary files, which reduces the risk of direct malware or automated RCE within the skill's own components.

ui_perfection