The Agent Skills Directory

[SAFE]: The skill implements a specialized persona for philosophical dialogue using standard markdown and persona instructions without malicious intent.
[SAFE]: Tool access is restricted to Read and AskUserQuestion, which limits the agent's capabilities to safe, interactive operations.
[SAFE]: External dependencies are restricted to local file references (../../framework.md and reference.md) for shared protocols and character context.
[SAFE]: No network exfiltration, hardcoded credentials, obfuscated content, or persistence mechanisms were detected in the skill files.

musashi