Establish AI agent environment policies, security guardrails, and team configuration standards.

• Configure project description files as AI manuals with tech stack, coding standards, and DO NOT rules; use /init for auto-generation from codebase analysis
• Set up security hooks to block dangerous commands (rm -rf, sudo, curl | sh) and auto-approve only safe operations via PreToolUse and PostToolUse events
• Define skills, slash commands, and plugins with token efficiency in mind; skills load on-demand while project descriptions load always
• Share team configurations via committed .claude/ folders and team-settings.json to maintain consistent standards across agents and team members
• Audit commands with cc-safe tool and use containerized environments for experimentation to enforce security policies