The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external configuration and data files (CARD, DATA) to perform statistical simulations. These represent ingestion points for untrusted data that could theoretically contain malicious instructions.
Ingestion points: Input files passed to the phack_cli.py script (e.g., DATA, CARD).
Boundary markers: None explicitly defined in the instructions.
Capability inventory: The agent is instructed to execute local CLI tools (python scripts/phack_cli.py) and use a Python library (phack).
Sanitization: Not described in the skill body; the agent relies on the logic within phack_cli.py and the phack module.

search-procedures