The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill includes functionality to install agent runtimes like OpenClaw and Hermes by downloading and executing shell scripts. This behavior is the intended purpose of the runtime management system and is protected by a multi-layer security check involving a regex-based injection guard and an optional AI-driven audit to prevent unauthorized command execution.- [EXTERNAL_DOWNLOADS]: The skill performs network requests to fetch installation scripts from the author's official GitHub repository and other well-known agent project sources. These downloads are part of the core functionality for setting up orchestration environments.- [SAFE]: Automated scanner alerts for reverse shell patterns and suspicious domains (e.g., evil.com) were found to be localized within the project's testing infrastructure (specifically src/lib/tests/injection-guard.test.ts). These are used as known-malicious samples to verify the efficacy of the application's built-in security filters and do not represent a threat to the user.

mission-control