The Agent Skills Directory

[SAFE]: The skill provides a structured framework for documentation and organization using a hub-and-spoke architecture. It contains no evidence of malicious intent, obfuscation, or unauthorized data access.\n- [COMMAND_EXECUTION]: The documentation guides users to execute local Python tools (e.g., scripts/analyze.py, scripts/tokens.py) and use chmod +x for permission management. These are standard developer operations for static code analysis and do not involve untrusted remote sources.\n- [PROMPT_INJECTION]: The modules/enforcement-patterns.md file describes templates for high-intensity prompt language (e.g., "YOU MUST", "NON-NEGOTIABLE"). These are intended as design patterns to help agents correctly prioritize specific skills in a complex environment and are not used as an attack against the agent's safety constraints.

modular-skills