The Agent Skills Directory

[SAFE]: The skill serves as a library of defensive security patterns and guidelines for agents to follow when building or auditing applications. It does not contain malicious code or override instructions.
[COMMAND_EXECUTION]: The included shell script scan-vulnerabilities.sh runs local auditing tools such as npm audit and bandit to identify security flaws in the project environment, which is the intended purpose of the skill.
[EXTERNAL_DOWNLOADS]: Guidelines for PII detection mention downloading official language models from Spacy's public repositories, which is a well-known and trusted technology service.
[SAFE]: The skill defines a platform-native PreToolUse hook that activates a dangerous command blocker to protect the execution environment during shell operations.

security-patterns