Analytical

Analytical MCP Server provides statistical analysis, decision support, logical reasoning, and research verification tools that run locally to assist Claude. It exposes MCP endpoints you can connect to from an MCP client to perform complex analyses and reasoning tasks.

How to use

You connect your MCP client to the Analytical MCP Server to perform statistical analyses, evaluate decisions with multi‑criteria scoring, inspect argument structures, and verify research claims. Start by running the server locally (or in Docker) and then configure your client to call the available tools through the MCP interface. Use the server to run data analyses on datasets you provide, generate visualization specifications, test hypotheses, and explore alternative perspectives on problems. For best results, keep your EXA API key secure and ensure it is available to the server when enabling research features.

How to install

Prerequisites: Ensure you have Node.js version 20.0.0 or higher and an EXA_API_KEY environment variable if you plan to use research features.

Option 1: Direct Installation to run locally

npm install
npm run build

Option 2: Docker to run in a container

# Build the Docker image
docker build -t analytical-mcp .

# Run with environment variables
docker run -d \
  --name analytical-mcp \
  -e EXA_API_KEY=your_api_key_here \
  -v $(pwd)/cache:/app/cache \
  analytical-mcp

# Or use docker-compose
cp .env.example .env
# Edit .env with your API key
docker-compose up -d

Configuration

Direct installation configuration includes a dedicated MCP server entry you place in your Claude Desktop settings to connect to the local node process.

{
  "mcpServers": {
    "analytical": {
      "command": "node",
      "args": ["/path/to/analytical-mcp/build/index.js"],
      "env": {
        "EXA_API_KEY": "your-exa-api-key-here"
      }
    }
  }
}

Docker configuration

If you prefer running the server through Docker, configure Claude Desktop to start the container with the proper environment and cache directory.

{
  "mcpServers": {
    "analytical": {
      "command": "docker",
      "args": [
        "run", "--rm", "-i",
        "--env-file", ".env",
        "-v", "$(pwd)/cache:/app/cache",
        "analytical-mcp"
      ]
    }
  }
}

Security and observability

All processing is designed to occur locally. Research features can use the EXA API key if you enable them. Cache data is stored locally and can be cleared as needed. You can monitor circuit breakers and cache metrics if you enable the metrics endpoint.

Enable metrics by setting the following environment variables to expose Prometheus‑style metrics at port 9090 by default.

METRICS_ENABLED=true        # Enable metrics server (default: true)
METRICS_PORT=9090          # Metrics server port (default: 9090)
METRICS_HOST=127.0.0.1     # Metrics server host (default: 127.0.0.1, use 0.0.0.0 to bind to all interfaces)

Usage examples

Use the available tools to analyze datasets, perform regression analyses, test hypotheses, or verify research claims. Example prompts focus on applying statistics to your data, understanding the relationships between variables, and validating arguments with structured reasoning.

Development notes

Typical development tasks include building, testing, and running an inspector for debugging.

Troubleshooting

If you encounter JSON parsing errors, direct your logs to stderr and ensure your MCP client reads stdout for responses. If tools do not appear, verify your CLI client configuration and restart the client. If research features are disabled, provide the EXA_API_KEY in your environment or .env file. If the server does not start, confirm you are using Node.js 20+ and that dependencies are installed with npm install.

Tool availability

The server exposes a range of tools for statistical analysis, decision analysis, logical reasoning, and research verification.

Available tools

analytical:analyze_dataset

Statistical analysis of datasets to uncover descriptive statistics, distributions, and relationships.

analytical:advanced_regression_analysis

Performs linear, polynomial, and logistic regression to model outcomes.

analytical:hypothesis_testing

Conducts t-tests, chi-square tests, and ANOVA to evaluate hypotheses.

analytical:data_visualization_generator

Generates specifications for data visualizations based on analysis results.

analytical:decision_analysis

Supports multi-criteria decision analysis with weighted scoring.

analytical:logical_argument_analyzer

Analyzes argument structure and assesses logical validity.

analytical:logical_fallacy_detector

Detects common fallacies in text to support critical evaluation.

analytical:perspective_shifter

Generates alternative perspectives on problems to broaden analysis.

analytical:verify_research

Cross-verifies research claims using multiple sources.