The Agent Skills Directory

[SAFE]: The skill serves as a defensive resource, instructing the agent to apply security constraints when handling user data, authentication, or external systems. It explicitly forbids dangerous practices such as using eval(), committing secrets to version control, or trusting client-side validation.
[SAFE]: All references within the skill, including the OWASP Top 10 quick reference, point to local files containing legitimate security documentation. No external network requests or remote code executions are initiated by the skill itself.
[SAFE]: The skill includes a verification checklist and clear categories for human-in-the-loop approval (e.g., changes to authentication flow), ensuring that high-risk operations are reviewed.

harden