The Agent Skills Directory

[SAFE]: No security threats were detected in the analyzed files. The skill functions as an advisory tool for privacy-preserving software development.
[REMOTE_CODE_EXECUTION]: The skill mentions and uses standard Python libraries for privacy and cryptography, such as opendp, numpy, and cryptography. These are used appropriately within code examples for educational purposes. No suspicious download-and-execute patterns or unverifiable dependencies were found.
[CREDENTIALS_UNSAFE]: The skill includes patterns to detect hardcoded secrets but does not contain any hardcoded credentials itself. It correctly advises using environment variables or secret management systems.
[DATA_EXFILTRATION]: No network operations or data exfiltration patterns were detected. The skill focuses on data minimization and privacy by design.
[INDIRECT_PROMPT_INJECTION]: While the skill processes user-provided content to offer privacy advice, it does not possess dangerous capabilities (such as file writes or network access) that could be exploited via injection. It also explicitly teaches PII detection and sanitization.

privacy-guardian