The Agent Skills Directory

[SAFE]: The skill functions as a legitimate setup and maintenance utility. It relies on bundled documentation and official vendor sources for prerequisites like Docker and Node.js.
[COMMAND_EXECUTION]: The skill uses various shell commands to manage the lifecycle of the KAIROS stack, including docker compose, npm install, and kairos CLI operations. All material actions, such as installing software or starting containers, are strictly gated behind explicit user confirmation.
[EXTERNAL_DOWNLOADS]: The skill facilitates the download of configuration files (e.g., compose.yaml) from the author's official GitHub repository (debian777/kairos-mcp). This is part of the intended functionality to support a no-clone installation path.
[CREDENTIALS_UNSAFE]: While the skill manages sensitive data such as OPENAI_API_KEY and authentication tokens, it implements defensive measures. It instructs the agent to store these in .env files or secure local configurations and explicitly forbids printing or repeating secret values in the user interface.

kairos-install