The Agent Skills Directory

[SAFE]: The skill serves as a comprehensive framework for security professionals to conduct AI/ML assessments. All logic provided is educational or serves as a template for security reporting and methodology planning.
[INDIRECT_PROMPT_INJECTION]: The skill defines an attack surface for processing system descriptions to generate threat models. While this is an ingestion point for untrusted data, it is a core functional requirement of a security analysis tool and the skill lacks the dangerous capabilities (like network exfiltration or sensitive file writes) necessary to facilitate an exploit. Evidence: Input schema in SKILL.md and conceptual analysis methods in AIThreatModel class.
[COMMAND_EXECUTION]: The script scripts/test-methodology.py is a utility used to generate a structured test plan. It uses only Python standard libraries and its operation is restricted to writing a local JSON file (test_methodology.json) containing the planned activities. No remote code execution or network operations are present.

testing-methodologies