The Agent Skills Directory

[SAFE] (SAFE): The skill follows security best practices, such as using yaml.safe_load() in its validation scripts and encouraging input validation in its design patterns.
[Indirect Prompt Injection] (LOW): The skill is designed to process external datasets, which presents a surface for indirect prompt injection.
Ingestion points: references/GUIDE.md identifies input_data as a primary entry point for processing.
Boundary markers: Absent; the skill does not define specific delimiters for separating data from instructions.
Capability inventory: scripts/validate.py uses os and pathlib for file system verification. The core ML libraries (PyTorch, TensorFlow) provide extensive computational capabilities.
Sanitization: references/PATTERNS.md suggests type and null checking, though it lacks specific sanitization for natural language inputs embedded in data.

Machine Learning