The Agent Skills Directory

[SAFE]: The skill is well-documented and focuses on a specific academic administrative task. It does not require any API keys, accounts, or network access, minimizing the attack surface.
[COMMAND_EXECUTION]: The skill instructs the agent to execute a local Python script scripts/aol_matrix.py using uv run python. Analysis of this script confirms it is a benign utility designed to validate the structure of curriculum matrices. It uses only standard Python libraries (csv, json, argparse) and does not perform any unsafe operations such as arbitrary code execution, subprocess spawning, or network communication.
[PROMPT_INJECTION]: The skill includes an indirect prompt injection surface because it processes untrusted user data (CSV files) and returns the parsed results to the agent.
Ingestion points: matrix.csv data read by scripts/aol_matrix.py.
Boundary markers: Absent; the agent is not explicitly told to ignore instructions found within the CSV cells.
Capability inventory: The skill has access to Read, Write, Edit, and Bash tools as defined in SKILL.md.
Sanitization: The script uses the standard csv.DictReader to parse data, which provides basic structural validation but does not filter content for malicious instructions.

alterlab-accreditation-aol