The Agent Skills Directory

[SAFE]: The skill provides legitimate documentation and architectural guidance for the ZooClaw Managed Agents platform. No malicious intent, obfuscation, or safety bypasses were detected within the instructions or referenced files.\n- [COMMAND_EXECUTION]: The skill documents the exec API, which allows running commands within the agent's sandbox environment. The documentation includes security-focused advice, such as instructing the AI to use argument arrays instead of shell strings to prevent command injection vulnerabilities in the resulting user applications.\n- [EXTERNAL_DOWNLOADS]: The skill references official vendor resources, including the zoowork-app-kit repository and developer documentation on GitHub (github.com/SerendipityOneInc). These resources are owned by the skill author, serendipityoneinc, and are used for legitimate setup and reference.\n- [SAFE]: The skill provides clear guidance on the secure handling of API keys, explicitly advising developers to store ZOOCLAW_API_KEY in server-side secret stores and never in client-side bundles or mobile apps.

zooclaw-managed-agents