The Agent Skills Directory

[SAFE]: The skill performs dynamic discovery of other components by scanning the 'src/skills' directory using 'Grep' and 'Read'. This is a standard documentation discovery pattern and is restricted to the local filesystem.
[SAFE]: Tool permissions are restricted to benign operations ('AskUserQuestion', 'Read', 'Grep', 'Glob'). No tools for network access, command execution, or system modification are requested in the frontmatter.
[SAFE]: The skill processes skill metadata from local files to build its listing. While this represents a surface for indirect instruction processing, the skill's lack of high-privilege capabilities (like code execution or network exfiltration) minimizes the risk.
[SAFE]: No obfuscation, hardcoded credentials, persistence mechanisms, or privilege escalation patterns were detected in the instructions or associated rule files.

help