The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes shell utilities (grep, glob) to traverse the file system and inspect Java source files within the ruoyi-modules directory.
[PROMPT_INJECTION]: The skill processes untrusted data from local source code, creating a surface for indirect prompt injection (Category 8).
Ingestion points: Text is read from *.java files via grep patterns targeting TODO: and FIXME: markers inside the ruoyi-modules/ directory.
Boundary markers: Content retrieved from files is interpolated into the agent's context without clear delimiters or 'ignore instructions' warnings.
Capability inventory: The skill uses grep and glob for data ingestion. It also references potential write capabilities through associated commands like /crud and /dev mentioned in the quick actions section.
Sanitization: No evidence of escaping or validation is present for the content extracted from code comments before it is processed by the AI.

progress