The Agent Skills Directory

[SAFE]: The skill serves as a static documentation and reference guide for web developers. It provides educational content on security headers, CSP, Trusted Types, and dependency management.
[COMMAND_EXECUTION]: The skill mentions command-line tools like npm audit, yarn audit, and openssl. These are standard development tools and are documented as recommendations for the user to perform security audits on their own projects, not as automated or hidden actions.
[EXTERNAL_DOWNLOADS]: The skill references well-known and trusted services such as Cloudflare (cdnjs), Mozilla, and MDN. These references are used to provide context and documentation for security features like Subresource Integrity (SRI) and polyfills. All such references are informative and do not involve untrusted remote code execution.
[PROMPT_INJECTION]: The skill includes an 'Anti-Patterns' section that explicitly instructs the agent to avoid treating external content or tool responses as trusted instructions, which is a defensive measure against prompt injection and autonomous abuse.

best-practices