amsl-schemas-mcp

# AMSL Schemas MCP Server A [Model Context Protocol (MCP)](https://modelcontextprotocol.io/) server for AMSL schema management, validation, and discovery. This server provides LLM agents with access to AMSL schema definitions and validation tools for YAML configuration files. ## Overview The AMSL Schemas MCP Server exposes AMSL schema resources and validation capabilities via the Model Context Protocol. It enables LLM agents to: - **Discover** available AMSL schema types (collection, assessment, module, agent, global, constants) - **Retrieve** full schema definitions from remote sources - **Validate** YAML data against AMSL schemas ## Features - 📚 **6 Schema Resources**: Dedicated resources for each AMSL schema type with detailed descriptions - ✅ **Validation Tool**: Validate data against any AMSL schema with detailed error reporting - 🔗 **Remote Schemas**: Automatically fetches schemas from the AMSL CDN - 🚀 **FastMCP**: Built with the FastMCP framework for minimal boilerplate - 📋 **Type Safety**: Full type hints and validation using Pydantic ## Supported Schemas - **Collection**: YAML structure for RAG document collections - **Assessment**: Assessment definitions for modules - **Module**: Module content structure with sessions and assessments - **Agent**: Agent definitions with steps and logic graphs - **Global**: Global configuration settings - **Constants**: Reusable constant values and prompts ## Installation ### From PyPI (once published) ```bash pip install amsl-schemas-mcp ``` ### From source ```bash git clone https://github.com/yourusername/amsl-schemas-mcp.git cd amsl-schemas-mcp pip install -e . ``` ## Quick Start ### Running the Server ```bash amsl-schemas-mcp ``` Or via Python module: ```bash python -m src.server ``` The server will start listening on stdio for MCP clients. ### Using with an LLM When connected to an LLM via MCP, you'll have access to: **Resources:** - `schema://amsl/collection` - Collection schema - `schema://…