The Agent Skills Directory

[SAFE]: The skill provides documentation and reference patterns for Clerk authentication in Next.js. Analysis of the 11 threat categories revealed no malicious behavior. Key findings include:
[DATA_EXPOSURE_AND_EXFILTRATION]: The skill defines required inputs for CLERK_SECRET_KEY and NEXT_PUBLIC_CLERK_PUBLISHABLE_KEY. These are standard requirements for Clerk integrations. No instructions or code patterns were found that attempt to exfiltrate these credentials or access sensitive local files.
[INDIRECT_PROMPT_INJECTION]: While the skill provides templates that the agent may use to generate code, these templates align with security best practices (e.g., using await auth().protect() or verifying userId in cache keys). The presence of boundary markers like callouts for different SDK versions helps ensure correct context.
[DYNAMIC_CONTEXT_INJECTION]: No !command syntax or load-time shell execution patterns were detected in the SKILL.md or other files.
[COMMAND_EXECUTION]: No suspicious subprocess calls, shell command execution, or script generation patterns were found. The code blocks provided are strictly for documentation and developer guidance.
[OBFUSCATION]: All content is in plain text. No Base64, zero-width characters, or homoglyph attacks were detected.

clerk-nextjs-patterns