The Agent Skills Directory

[SAFE]: The skill is limited to read-only operations using the Grep, Glob, and Read tools. It does not possess capabilities for file modification, network access, or command execution.
[SAFE]: The documentation includes security best practices for the agent, such as validating regex syntax and rejecting path traversal patterns (e.g., '..') and null bytes in search inputs.
[SAFE]: No obfuscation, hardcoded credentials, or unauthorized external dependencies were detected in the skill definition.

code-search