The Agent Skills Directory

[SAFE]: The skill provides educational content and templates for security design reviews based on established industry frameworks like STRIDE and Adam Shostack's methodology.
[DATA_EXPOSURE]: While the text mentions sensitive locations such as .env files, S3 buckets, and PII, it does so strictly in the context of identifying potential threats during a modeling session, not as an attempt to access or exfiltrate real data.
[REMOTE_CODE_EXECUTION]: The skill contains no scripts, package installations, or remote code download patterns. It relies entirely on standard instructional text.
[PROMPT_INJECTION]: The instructions are focused on the task of threat modeling and do not contain any patterns intended to bypass AI safety guidelines or override system behavior.
[COMMAND_EXECUTION]: No shell commands or subprocess calls are present in the skill instructions or examples.

threat-modeling