The Agent Skills Directory

[SAFE]: The skill provides structural guidance and best practices for project management without introducing security vulnerabilities.
[DYNAMIC_EXECUTION]: The provided Python scripts (scripts/create_subagent.py and scripts/generate_claude_md.py) are local utility tools that generate Markdown templates and JSON configurations on the local file system. These scripts do not download remote code or execute arbitrary system commands from external sources.
[COMMAND_EXECUTION]: While the skill references CLI commands like /agent, /clear, and /compact, these are built-in platform features of the Claude Code environment and are used as intended for session management.
[DATA_EXPOSURE]: No hardcoded credentials or access to sensitive file paths (e.g., SSH keys, AWS credentials) were found. The skill correctly identifies .env files as a location for secrets but does not attempt to read or leak them.
[PROMPT_INJECTION]: The instructions use strong language (e.g., 'STOP', 'MANDATORY', 'CRITICAL') to enforce development best practices such as planning before writing and using correct file paths on Windows. These are standard instructional guidelines and do not attempt to bypass AI safety filters.

context-master