The Agent Skills Directory

[SAFE]: The skill serves as a comprehensive reference for the litestar-mcp library, which is designed to bridge Litestar applications with the Model Context Protocol.
It provides architectural documentation for stateless MCP transport over HTTP and stdio.
It includes decorators and controllers for exposing tools, prompts, and resources to AI agents.
[EXTERNAL_DOWNLOADS]: Installation instructions reference official packages from public registries and well-known services.
It lists dependencies such as litestar-mcp, httpx2, a2a-sdk, and google-genai which are required for the framework's core functionality and optional extensions.
[INDIRECT_PROMPT_INJECTION]: The skill describes a framework for handling external input from AI models, which constitutes a potential attack surface.
The documentation provides guidance on mitigating risks by using litestar-security for authentication and authorization policies, as well as Litestar guards to restrict access to sensitive tools and handlers.

litestar-mcp