MCP Solver

Provides constraint solving, SAT, SMT, and ASP capabilities via MCP for interactive model creation and solving.

python

0

GitHub Stars

python

Language

5 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "swesmith-repos-szeider__mcp-solver.c431fabc": {
      "command": "mcp-solver-mzn",
      "args": []
    }
  }
}

You run the MCP Solver server to access constraint solving, SAT, SMT, and ASP capabilities from large language models. It provides five backends (MiniZinc, PySAT, MaxSAT, Z3, and ASP) that you can use through an MCP client to model and solve problems interactively with an LLM.

How to use

To use the MCP Solver, run the MCP client and connect to one of the available backends. You can switch modes to work with MiniZinc, PySAT, MaxSAT, Z3, or ASP by selecting the corresponding backend through the client. The client translates natural language problem statements into formal constraint models and communicates with the server to build, edit, and solve those models.

How to install

Prerequisites: you need Python 3.11 or later and the uv package manager.

Step 1: Clone the project repository.

git clone https://github.com/szeider/mcp-solver.git
cd mcp-solver

Step 2: Create and activate a virtual environment using uv.

uv venv
source .venv/bin/activate

Step 3: Install all solver backends and client dependencies.

uv pip install -e ".[all]"  # Install all solvers
uv pip install -e ".[client]"  # If you plan to use the MCP Test Client

Additional sections

MCP Solver supports five solving backends, each exposed as a distinct run configuration you can invoke with the MCP client.

Notes and tips

The server provides dedicated modes for each backend: MiniZinc, PySAT, MaxSAT, Z3, and ASP. Use the appropriate mode command to start the corresponding backend after installation.

Available tools

clear_model

Remove all items from the current model.

add_item

Add new item at a specific index in the model.

delete_item

Delete the item at a given index from the model.

replace_item

Replace the item at a specific index with new content.

get_model

Retrieve the current model content with numbered items for inspection.

solve_model

Solve the model, optionally with a timeout, and return the solution.