The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses shell commands to manage the installation and verification lifecycle.
Executes docker compose to start and manage containerized services for the KAIROS stack.
Uses npm and npx for managing the CLI and related tools.
Employs network utility commands such as curl, lsof, ss, and netstat to perform health checks and identify port conflicts before starting services.
[EXTERNAL_DOWNLOADS]: Fetches configuration and installation resources from remote sources.
Downloads the compose.yaml file from the author's official GitHub repository (debian777/kairos-mcp).
References official download pages and documentation for well-known services including Docker, Node.js, and Ollama to assist the user with prerequisite installation.
[REMOTE_CODE_EXECUTION]: Facilitates the installation and execution of software components from the vendor's repository.
Installs the @debian777/kairos-mcp CLI package via npm, which is the primary interface for the software.
Supports the dynamic addition of related agent skills using the npx skills add debian777/kairos-mcp command.
[PROMPT_INJECTION]: Implements defensive instructions to maintain agent integrity and user safety.
Explicitly forbids the agent from asking users to provide secrets (API keys, tokens) in chat or echoing them back in the output.
Commands the agent to prioritize bundled reference documentation over potentially untrusted remote content, reducing the risk of indirect prompt injection from upstream sources.

kairos-install