The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to resolve file paths using glob expansion and to execute grep for cross-file pattern detection, such as identifying symbols called from only one file.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and scans the full contents of untrusted source code files provided by the user. Ingestion points: Local source code files resolved via glob patterns or paths in SKILL.md. Boundary markers: Absent; the instructions do not include delimiters or warnings to ignore embedded instructions within the source files. Capability inventory: The agent has the capability to read file contents and execute search commands like grep. Sanitization: Absent; the skill performs analysis on raw file content without sanitization or input validation.

anti-slop