- Home
- MCP servers
- DART
DART
- 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": {
"2geonhyup-dart-mcp-test": {
"command": "uv",
"args": [
"--directory",
"/Users/{컴퓨터이름}/Downloads/dart-mcp",
"run",
"dart.py"
],
"env": {
"DART_API_KEY": "{DART_API_KEY}"
}
}
}
}You can extend Claude with DART-MCP to analyze and visualize corporate financial data through the DART API. This MCP connects Claude to the DART data source, enabling structured financial insights, detailed analyses, and visualizations directly in your conversations.
How to use
You use a Claude desktop client configured to run the DART-MCP server locally. After you set up the integration, ask Claude to fetch data from the DART API and perform financial analyses. You can request major financial trends, department-level sales breakdowns, and valuation-oriented metrics such as discounted cash flow. Claude will query the DART API and return analyzed results, including visualizations created via Claude. If you need to adjust inputs, specify the company name using its official listing name and request quarterly or annual figures as needed.
How to install
{
"mcpServers": {
"dart_mcp": {
"command": "uv",
"args": ["--directory", "/Users/{컴퓨터이름}/Downloads/dart-mcp", "run", "dart.py"],
"env": {
"DART_API_KEY": "{DART_API_KEY}"
}
}
}
}
Prerequisites
- Ensure you have a DART API key
- Install Claude desktop app and sign in
- Have a working internet connection for API calls
Step-by-step
- Download the MCP package from the project repository
- Extract the ZIP file and verify the folder sits in your Downloads directory
- Rename the extracted folder to dart_mcp if needed (keep the name exact if it already is dart_mcp)
- Open the Claude desktop app
- Access settings to editing mode
- Mac: Claude > Settings > Developer > Edit settings
- Windows: Settings > Developer > Edit settings
- Open the top-level claude_desktop_config file in a text editor
- Replace placeholders with your values:
- DART_API_KEY: your issued API key
- {컴퓨터이름}: your computer account name (Mac: Finder home; Windows: your user folder name)
- Save the configuration file
- Restart Claude to apply the new MCP configuration
- Start asking Claude questions that involve DART data to receive analyzed responses and visualizations
## Configuration and usage notes
Only official company names (as listed) should be used when querying data. The MCP currently targets KOSPI/KOSDAQ listings. Real-time price data or market capitalization may be added in future updates, but are not part of the current MCP capabilities. You should provide a valid DART API key and ensure Claude can reach the DART API endpoints.
## Examples of capabilities
- Major financial analysis: summarize revenue and operating income trends by quarter or year for a given company.
- Detailed financial analysis: drill into line items, margins, and drivers behind variances.
- Segment/Business unit analysis: breakdown of revenue by business segment or subsidiary where available.
- Visualization: generate charts and graphs to illustrate revenue, profit, or cash flow trends.
- Valuation support: apply financial metrics such as DCF using data from the DART API for informed assessments.
## Security and usage cautions
Keep your DART API key secure and do not share it in public channels. Use the MCP only with officially listed company names and verify the scope of data you request. Remember that financial data may have limitations or delays depending on the source.
## Available tools
### financial\_analysis
Perform major financial analysis including revenue and operating income trends and key drivers.
### visualization
Create charts/graphs to visualize financial data such as revenue by segment or time series trends.
### valuation\_metrics
Provide valuation-related insights using metrics like DCF based on DART data.