Quantitative Research

The Quantitative Research MCP Server helps you organize and track quantitative analyses within a structured knowledge graph. It supports persistent research contexts, session management, hypothesis testing, dataset and variable tracking, statistical analyses, model performance, visualizations, and documented research findings. This makes it easy to connect data, analyses, and conclusions across iterative research efforts.

How to use

You interact with the Quantitative Research MCP Server through a client that communicates with the MCP backend. Start a new session to generate a unique session ID and begin working on your Climate Impact Study or any project you manage. Load context for specific entities to review their current status, priority, and sequence relations. Record analyses, create visualizations, and link findings to supporting statistical evidence. End a session by summarizing what you did, updating datasets and models, and documenting status changes to keep your research lifecycle organized.

Key capabilities you will use often include creating entities (projects, datasets, variables, hypotheses, tests, results, models, visualizations, findings), establishing relationships (correlates_with, predicts, tests, contains, part_of, supports, cites, has_status, has_priority, precedes), and retrieving domain-specific views (project overviews, dataset analyses, hypothesis test results, variable relationships, model performance). You can also document methodology decisions and track research questions to ensure a rigorous, auditable workflow.

How to install

Prerequisites: you need Node.js and npm installed on your system. You can verify by running node -v and npm -v in your terminal. You also have the option to run the MCP server via a remote or container-based approach if you prefer not to install locally.

Option A: Run via npm/npx (recommended for quick setup) You can start by configuring your MCP client to use the npx-based runtime. The configuration snippet shown below adds the MCP server to your client:

{
  "mcpServers": {
    "quantitativeresearch": {
      "command": "npx",
      "args": [
        "-y",
        "github:tejpalvirk/quantitativeresearch"
      ]
    }
  }
}

How to install

Option B: Install globally and run directly If you prefer a global installation, run the following command to install the package and then configure your client accordingly:

npm install -g github:tejpalvirk/quantitativeresearch

{
  "mcpServers": {
    "quantitativeresearch": {
      "command": "contextmanager-quantitativeresearch"
    }
  }
}

How to install

Option C: Run via Docker If you prefer containerized execution, you can start with Docker using the following command:

{
  "mcpServers": {
    "quantitativeresearch": {
      "command": "docker",
      "args": [
        "run",
        "--rm",
        "-i",
        "mcp/quantitativeresearch"
      ]
    }
  }
}

Configuration notes

Environment customization lets you define where data and session information are stored. Two variables control this behavior: MEMORY_FILE_PATH and SESSIONS_FILE_PATH. You can set defaults to ensure data persists across sessions, for example: MEMORY_FILE_PATH="./quantitativeresearch/memory.json" and SESSIONS_FILE_PATH="./quantitativeresearch/sessions.json".

Additional setup and tips

If you are building from source, you may proceed with a local build, followed by running the server. Typical steps include cloning the project, installing dependencies, building, and starting the server with the appropriate entry script. Docker users can also build a container image as shown in the Docker instructions.

Available tools

getProjectOverview

Provides a comprehensive view of a project including research questions, methodology, datasets, and variables.

getDatasetAnalysis

Generates analysis of dataset contents including variables, descriptive statistics, and data quality.

getHypothesisTests

Retrieves hypothesis tests and their outcomes for a given study.

getVariableRelationships

Examines correlations, predictions, and other relationships between variables.

getStatisticalResults

Summarizes results of statistical analyses.

getVisualizationGallery

Lists visualizations created for datasets and results.

getModelPerformance

Assesses performance metrics for statistical models.

getResearchQuestionResults

Organizes analyses and results by research questions.

getVariableDistribution

Examines distribution and properties of individual variables.

getStatusOverview

Views all entities with a specific status (active/completed/pending/abandoned).

getPriorityItems

Identifies high-priority research tasks and activities.

getResearchSequence

Visualizes the sequence of research processes based on precedes relations.