The Agent Skills Directory

[SAFE]: The skill contains no executable code, shell scripts, or commands that interact with the host system. It is entirely prompt-based and operates on data provided within the chat context.
[SAFE]: The instructions explicitly direct the agent to prioritize security and privacy during code reviews, including checks for exposed secrets, unsafe input handling, and authorization gaps.
[SAFE]: No obfuscation, data exfiltration, or persistence mechanisms are present. The skill uses a structured output format for clarity and educational purposes.
[SAFE]: References to specialist skills (e.g., $golang-pro, $supabase-postgres-best-practices) are used for context-aware consultation and do not involve downloading external or untrusted code.

sensei-spar