The Agent Skills Directory

[SAFE]: The skill facilitates codebase navigation and analysis using standard search and version control tools. No malicious behaviors, obfuscation, or unauthorized access attempts were detected.
[DATA_EXFILTRATION]: The instructions include patterns for identifying configuration files and environment variables (e.g., searching for 'API_KEY' or '.env'). While this targets sensitive data, it is a standard developer activity for auditing and configuring applications and does not include mechanisms for exfiltration.
[PROMPT_INJECTION]: The skill processes external codebase content which could contain adversarial instructions. Ingestion points: Files are read using Read, Grep, and Glob tools; Boundary markers: None; Capability inventory: Bash, Read, Grep, Glob tools are accessible; Sanitization: None.

codebase-search