The Agent Skills Directory

[SAFE]: The skill serves as an educational resource and policy template for hardening AI agent environments. It includes defensive configurations such as hooks to block destructive commands like 'rm -rf' and guidelines for preventing secret exposure.
[COMMAND_EXECUTION]: Mentions the use of environment-specific commands like '/init' and '/sandbox' for project management. These are documented as standard operations within the supported agent frameworks for providing guardrails.
[EXTERNAL_DOWNLOADS]: Recommends the use of 'npx cc-safe' for auditing projects. This is presented as a manual, user-initiated security workflow for code analysis and is consistent with the skill's primary purpose as a security guide.

agent-configuration