mcp-llm-generator

The MCP LLM Generator v2 server provides a production-ready MCP (Model Context Protocol) server focused on advanced large language model generation, memory-enabled interactions, and a rich template/prompt ecosystem. Its Core Tools include llm-generate for direct LLM text generation via the MCP sampling protocol, template-execute for sophisticated parameterized templates, template-manage for CRUD operations on reusable prompts, context-chat for personality-driven conversations with persistent memory, context-manage for creating and managing AI consultant contexts, and memory-store to organize knowledge with associative linking. Experimental features (v1.3.0+) add self-awareness and cross-awareness capabilities, capability inheritance, and capability matrices across multiple personas. To use the server, install the MCP LLM Generator CLI globally or run it from your project, and then leverage client calls to the available tools to generate text, manage templates, chat with context-aware consultants, and store memory. See the Quick Start and Usage Examples for concrete tool invocations in both production and development configurations.

Prerequisites:

• Node.js (LTS) and npm installed on your machine
• Basic familiarity with MCP client configuration

Global installation (recommended for production):

# Install globally for enterprise deployment
npm install -g @mako10k/mcp-llm-generator

# Verify installation and health checks
mcp-llm-generator --version
mcp-llm-generator --health-check

Local development setup (Sprint3 Enhanced):

# Clone the repository with Sprint3 development context
git clone https://github.com/mako10k/mcp-sampler.git
cd mcp-sampler

# Install dependencies with integrity check
npm ci

# Run development server with Sprint3 tooling
npm run dev
./scripts/dod-check.sh          # Quality gate validation
./scripts/mcp-integration-test.sh  # MCP protocol testing

Optional: validate client configuration (example blueprint):

npx @modelcontextprotocol/inspector node build/index.js

For production deployment, install the CLI globally and use the provided mcp-llm-generator executable as your server entry point.

Notes:

• The project emphasizes Sprint3 process quality with automated DoD checks, MCP protocol validation, and CI/CD gates.
• Experimental features (Step3 capability awareness) are under active development and may change; test accordingly before relying on them in production.
• The MCP client configurations show both production (global) and development (local) JSON formats. Use the one that matches your deployment strategy.
• If using templates or memory operations, ensure proper memory governance and security considerations for sensitive data.
• Health checks and build/test scripts are included in the repository to maintain quality gates during deployment.
• Environment variables like LOG_LEVEL and NODE_ENV are commonly used in client configurations to control verbosity and environment behavior.