mcp-solver
The MCP Solver is a server aimed at integrating SAT, SMT, and constraint solving with large language models using the Model Context Protocol. It offers interactive capabilities for creating, editing, and solving models with three different operational modes. This includes extensive support for various problem-solving backends and a test client for experiments.
What is the MCP Solver?
The MCP Solver is a server that integrates SAT, SMT, and Constraint Solving capabilities with LLMs, allowing for interactive model creation, editing, and solving.
What are the system requirements for MCP Solver?
MCP Solver requires Python 3.11+, the 'uv' package manager, and solver-specific dependencies like MiniZinc, Z3, or PySAT.
How do I install the MCP Solver?
Clone the repository, create a virtual environment using 'uv', activate it, and install all solvers using 'uv pip install -e "[all]"'.
What operational modes does MCP Solver support?
MCP Solver supports MiniZinc, PySAT, and Z3 modes, each with specific dependencies and capabilities.
Can I use MCP Solver with different LLM providers?
Yes, the MCP client supports various LLM providers including Anthropic, OpenAI, Google, OpenRouter, and local models via LM Studio.