The Agent Skills Directory

[SAFE]: The skill implements a security validation workflow that uses standard patterns to detect Common Weakness Enumerations (CWE) such as SQL injection, XSS, and hardcoded credentials. It functions as a pre-commit security gate to improve code quality.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it is designed to process untrusted code files during the development lifecycle.
Ingestion points: The skill triggers on modifications to sensitive project areas, including **/auth/**, **/api/**, and .env files.
Boundary markers: The instructions do not specify explicit delimiters or "ignore" instructions for the agent when processing the content of these files.
Capability inventory: The skill identifies vulnerabilities using pattern matching and provides documentation for auto-fixes. It does not include executable scripts, network operations, or arbitrary command execution capabilities.
Sanitization: The skill does not perform sanitization of the input code; however, its primary purpose is to identify and report on those very inputs to the user.

security-pipeline