The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface through the ingestion of external resume files in multiple formats (DOCX, PDF, HTML, MD, TXT). \n
Ingestion points: User-provided document files are read and processed by scripts/parse_resume.py. \n
Boundary markers: The skill does not provide the agent with specific delimiters or instructions to ignore potential commands embedded within the resume text. \n
Capability inventory: The skill utilizes Read, Write, Edit, Bash, Glob, and Grep tools, which define its operational scope. \n
Sanitization: Extracted text is passed to the agent without specific filtering or sanitization for malicious instructions. \n- [COMMAND_EXECUTION]: The use of subprocess.run in tests/test_init_profile.py is identified as a benign implementation for validating the skill's own CLI tools during the testing phase. \n- [EXTERNAL_DOWNLOADS]: The skill declares dependencies on python-docx and pypdf using PEP 723 metadata in scripts/parse_resume.py. These are well-established libraries from reputable sources.

career-profile