The Agent Skills Directory

[SAFE]: The skill provides defensive security review guidelines and education. All code examples in the references are clearly labeled as either vulnerable or secure, serving as a knowledge base for the agent. The permitted tools (Read, Edit, Grep, Glob, Bash) are used strictly for identifying security-sensitive patterns within a codebase, such as searching for hardcoded secrets or assessing route configurations. No evidence of prompt injection, unauthorized data access, or malicious persistence was found across the skill's instructions or its reference documentation.

typescript-security-review