home / mcp / dart mcp server

DART MCP Server

Provides financial data analysis and visualization through DART API access via Claude.

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "2geonhyup-dart-mcp": {
      "command": "uv",
      "args": [
        "--directory",
        "/Users/{computer_name}/Downloads/dart-mcp",
        "run",
        "dart.py"
      ],
      "env": {
        "DART_API_KEY": "YOUR_API_KEY"
      }
    }
  }
}

You can leverage the DART MCP server to perform model-assisted financial analysis using Claude. It connects to the DART API to fetch company data, enables visualization, and supports valuation analyses. This MCP setup lets you run queries against the local Claude client to obtain structured financial insights and charts for listed companies.

How to use

Start Claude and ensure the DART MCP server is properly configured in the Claude desktop app. When you ask Claude about a company’s financials or comparisons, your queries will be routed through the DART API to fetch data, which Claude will then visualize or summarize for you. Use specific company names (officially listed names) and target KOSPI/KOSDAQ stocks. You can request quarterly revenue, operating income trends, segment sales by business unit, and valuation insights that rely on financial metrics.

Examples of useful queries include analyzing revenue and operating income trends by quarter, comparing profitability metrics between two companies for a given year, and evaluating debt or risk indicators for a given issuer. If you need visualizations, ask Claude to generate charts for the requested data so you can interpret trends more easily.

How to install

Prerequisites you need before starting: Node or Python tools may be used by the MCP, and Claude desktop app must be installed on your computer. Prepare a DART API key to authorize requests.

Follow these concrete steps to set up the MCP server integration with Claude on your machine.

Additional sections

Configuration details from the MCP setup are shown below. This configuration enables Claude to call the DART API for financial data analysis.

{
  "mcpServers": {
    "dart-mcp": {
      "command": "uv",
      "args": ["--directory", "/Users/{컴퓨터이름}/Downloads/dart-mcp", "run", "dart.py"],
      "env": {
        "DART_API_KEY": "{DART_API_KEY}"
      }
    }
  }
}

Available tools

financial_analysis

Performs key financial analyses such as revenue, profit trends, and margin analysis across time periods.

visualization

Generates visual representations of financial data, including graphs for revenues, operating income, and segment sales by business unit.

company_comparison

Compares profitability and efficiency metrics between companies, aiding competitive assessments.

risk_assessment

Evaluates financial risk by examining leverage, debt levels, and balance sheet structure.