The Agent Skills Directory

[COMMAND_EXECUTION]: The skill constructs shell commands by interpolating a file path directly into a command string for execution via python3 or textutil. This is a classic command injection vulnerability where a maliciously named file (e.g., one containing semicolons or backticks) could execute arbitrary code when processed.
Evidence: python3 -c "import PyPDF2; ... open('FILE_PATH','rb') ... " and textutil -convert txt -stdout "FILE_PATH".
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and processes untrusted data from external files.
Ingestion points: Reads CV/resume files identified via glob patterns in the current working directory.
Boundary markers: Absent. The skill does not use delimiters or instructions to ignore embedded commands within the CV content.
Capability inventory: The skill can execute shell commands (python3, textutil) and write to the local filesystem (.local.md).
Sanitization: Absent. The extracted content is processed and analyzed directly without any validation or filtering.

CV Analysis