The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to execute mkdir -p for creating the directory structure of new philosopher personas. This is a legitimate administrative task within the skill's defined scope and is used only for local file management.
[INDIRECT_PROMPT_INJECTION]: The skill functions as a meta-generator, creating new instruction files from user input and existing local templates. Ingestion points: Local template files (e.g., framework.md, nietzsche/SKILL.md) and the user-provided <name> argument. Boundary markers: None explicitly defined for read operations. Capability inventory: File system access via Read, Write, and Bash tools. Sanitization: Not explicitly specified in instructions. This attack surface is mitigated by the mandatory 'Review Gate' (Step 3), which requires the agent to present generated content for user verification before it is written to the file system.
[SAFE]: No malicious patterns, hardcoded credentials, obfuscation, or unauthorized data exfiltration attempts were detected. The skill's behavior is consistent with its stated purpose of persona generation and library management.

create