The Agent Skills Directory

[SAFE]: The skill implements standard project management practices for Claude Code environments, focusing on project context, coding standards, and persistent memory management.
[DATA_EXPOSURE]: The skill manages configuration files in standard locations (e.g., ~/.claude/CLAUDE.md). It explicitly warns users against including sensitive data like credentials or API keys and provides templates illustrating secure practices (using environment variables instead of hardcoded secrets).
[INDIRECT_PROMPT_INJECTION]: The skill's primary function is to create and modify instruction files that influence future agent behavior. It includes a 'MANDATORY Validation' section that requires the agent to verify content is non-sensitive and appropriate for persistent memory before writing, which acts as a safeguard against misusing the instruction-loading mechanism.

edit-claude