Skip to main content
Glama

DART-MCP: Claude Extension for Financial Analysis

This is a financial analysis MCP (Model-assisted Capability Package) utilizing DART API. Using Claude, you can easily analyze and visualize financial data of listed companies.

I'll add a more detailed and easy guide at the top, which you can see at https://dart-mcp.vercel.app/ .

Possible / Impossible

Possible (O)

  • Key Financial Analysis

  • Detailed financial analysis

  • Sales by business unit of the company

  • Visualization using Claude

  • Valuation using financial indicators (DCF, etc.)

Impossible (X)

  • Provides stock price and market capitalization

  • Analysis of overseas companies

  • Use more than Claude's free usage

  • Use multiple chat windows (if it doesn't work, create a new chat window and use it)

  • 100% accurate information

The investment information provided may differ from the actual information, and investment responsibility lies with the investor.

Related MCP server: AuthMCP Gateway

Example of use

Financial data analysis and visualization

파마리서치의 2023, 2024년 매출액, 영업이익 추이 분기별로 그래프로 보여줘. 그리고 매출비중이 어떻게 되는지 알려줘. 영업이익이나 매출액 변동 이유도 분석해줘.

Comparative Analysis of Companies

카카오와 네이버 2024년 수익성지표를 비교해서 분기별로 보여주고, 각 기업들은 어떤 사업부가 성장을 이끌지 알려줘.

Financial Risk Assessment

한국전력의 최근 부채상황을 조사하고, 상세하게 어떤 부분이 문제인지 분석해줘.

Advance preparation

Issuing DART API Key

  1. Access the DART Open API website

  2. Sign up and log in

  3. [Authentication Key Application/Management] - Click [Open API Use Application] menu

  4. Apply after entering usage information

  5. [Authentication Key Application/Management] - Check the issued authentication key in the [Open API Usage Status] menu

Install the Claude desktop app

  1. Download the Claude desktop app

  2. Sign up and log in to your account

How to install

1. Download the project from GitHub

Download the zip file from the GitHub page: https://github.com/2geonhyup/dart-mcp

2. Unzip the ZIP file and check the folder location.

  1. Unzip the downloaded ZIP file.

  2. Make sure the unzipped folder is in Downloads . If it's somewhere else, move it to the Downloads location.

3. Change the folder name

Be sure to change the name of the unzipped folder dart-mcp-main to dart-mcp. (If it was dart-mcp from the beginning, don't change it.)

4. Access the Claude app and settings

  1. Launch the installed Claude desktop app.

  2. Mac users: Click Claude > Settings > Developer > Edit Settings Windows users: Click Settings > Developer > Edit Settings

5. Open the settings file

Open the claude_desktop_config file at the top with a text editor.

6. Enter the setup code

  1. First, enter the correct key and name.

    • DART API Key: Enter the API key you received

    • Computer Name: Enter your computer account name (Finder home folder on Mac, C:\Users folder name on Windows)

  2. Enter the following code into the settings file:

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

7. Restart Claude and start using it

Save the config file, close the Claude app, and restart it . Now when you ask Claude a question, it will call the DART API and provide the answer.

Precautions for use

  • The company name must be provided as its officially listed name.

  • Only KOSPI and KOSDAQ stocks can be researched.

  • We plan to link real-time information such as stock prices and market capitalization in the future.

Available Tools

5 tools
get_current_dateA
현재 날짜를 YYYYMMDD 형식으로 반환하는 도구

Args:
    ctx: MCP Context 객체 (선택 사항)
    
Returns:
    YYYYMMDD 형식의 현재 날짜 문자열

Inputs: no named parameters.

search_business_informationB
회사의 사업 관련 현황 정보를 제공하는 도구

Args:
    company_name: 회사명 (예: 삼성전자, 네이버 등)
    start_date: 시작일 (YYYYMMDD 형식, 예: 20230101)
    end_date: 종료일 (YYYYMMDD 형식, 예: 20231231)
    information_type: 조회할 정보 유형 
        '사업의 개요' - 회사의 전반적인 사업 내용
        '주요 제품 및 서비스' - 회사의 주요 제품과 서비스 정보
        '원재료 및 생산설비' - 원재료 조달 및 생산 설비 현황
        '매출 및 수주상황' - 매출과 수주 현황 정보
        '위험관리 및 파생거래' - 리스크 관리 방안 및 파생상품 거래 정보
        '주요계약 및 연구개발활동' - 주요 계약 현황 및 R&D 활동
        '기타 참고사항' - 기타 사업 관련 참고 정보
    ctx: MCP Context 객체
    
Returns:
    요청한 정보 유형에 대한 해당 회사의 사업 정보 텍스트

Inputs: company_name (required), start_date (required), end_date (required), information_type (required).

search_detailed_financial_dataB
회사의 세부적인 재무 정보를 제공하는 도구.
XBRL 파일을 파싱하여 상세한 재무 데이터를 추출합니다.

Args:
    company_name: 회사명 (예: 삼성전자, 네이버 등)
    start_date: 시작일 (YYYYMMDD 형식, 예: 20230101)
    end_date: 종료일 (YYYYMMDD 형식, 예: 20231231)
    ctx: MCP Context 객체
    statement_type: 재무제표 유형 ("재무상태표", "손익계산서", "현금흐름표" 중 하나 또는 None)
                   None인 경우 모든 유형의 재무제표 정보를 반환합니다.
    
Returns:
    선택한 재무제표 유형(들)의 세부 항목 정보가 포함된 텍스트

Inputs: company_name (required), start_date (required), end_date (required), statement_type.

search_disclosureA
회사의 주요 재무 정보를 검색하여 제공하는 도구.
requested_items가 주어지면 해당 항목 관련 데이터가 있는 공시만 필터링합니다.

Args:
    company_name: 회사명 (예: 삼성전자, 네이버 등)
    start_date: 시작일 (YYYYMMDD 형식, 예: 20230101)
    end_date: 종료일 (YYYYMMDD 형식, 예: 20231231)
    ctx: MCP Context 객체
    requested_items: 사용자가 요청한 재무 항목 이름 리스트 (예: ["매출액", "영업이익"]). None이면 모든 주요 항목을 대상으로 함. 사용 가능한 항목: 매출액, 영업이익, 당기순이익, 영업활동 현금흐름, 투자활동 현금흐름, 재무활동 현금흐름, 자산총계, 부채총계, 자본총계
    
Returns:
    검색된 각 공시의 주요 재무 정보 요약 텍스트 (요청 항목 관련 데이터가 있는 경우만)

Inputs: company_name (required), start_date (required), end_date (required), requested_items.

search_json_financial_dataA
회사의 재무 정보를 JSON API를 통해 제공하는 실패시 보완하는 보조 도구.
search_disclosure, search_detailed_financial_data이 2023년 9월 이전 자료 분석에 실패했을 때 대안으로 활용.

Args:
    company_name: 회사명 (예: 삼성전자, 네이버 등)
    bsns_year: 사업연도 (4자리, 예: "2023")
    ctx: MCP Context 객체
    reprt_code: 보고서 코드 ("11011": 사업보고서, "11012": 반기보고서, "11013": 1분기보고서, "11014": 3분기보고서)
    fs_div: 개별/연결구분 ("OFS": 재무제표, "CFS": 연결재무제표)
    statement_type: 재무제표 유형 ("BS": 재무상태표, "IS": 손익계산서, "CIS": 포괄손익계산서, "CF": 현금흐름표, "SCE": 자본변동표)
                   None인 경우 모든 유형의 재무제표 정보를 반환합니다.
    
Returns:
    선택한 재무제표 유형(들)의 세부 항목 정보가 포함된 텍스트 (당기 데이터만 표시)

Inputs: company_name (required), bsns_year (required), reprt_code, fs_div, +1 more.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.6/5.0

    Scored across 5 tools

    Disambiguation4/5

    Most tools have distinct purposes: get_current_date is a simple utility, while the other four focus on company information retrieval with clear specializations. However, search_detailed_financial_data and search_json_financial_data both provide financial data with overlapping functionality, creating some potential confusion despite different data sources and parameter structures.

    Naming Consistency5/5

    All five tools follow a consistent snake_case naming pattern with clear verb_noun structure. Each name accurately describes the tool's function (get_current_date, search_business_information, search_detailed_financial_data, search_disclosure, search_json_financial_data), maintaining excellent consistency throughout the set.

    Tool Count4/5

    Five tools is reasonable for a financial/business data server, though the scope feels slightly thin. The set covers date retrieval and company information with multiple financial data approaches, but could potentially benefit from additional complementary tools like company listing or sector analysis to feel more complete.

    Completeness3/5

    For a Korean financial data server, there are notable gaps in coverage. While financial data retrieval is well-represented with multiple approaches, there's no tool for listing available companies, searching by sector/industry, retrieving historical stock prices, or accessing corporate governance information. The server focuses heavily on financial statements and disclosures but misses broader corporate data aspects.

    Maintenance

    ActivityInactive
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      A
      maintenance
      txtai is an all-in-one embeddings database for semantic search, LLM orchestration and language model workflows. All functionality can be served via it's API and the API supports MCP. Docs: https://neuml.github.io/txtai/api/mcp/
      13,377
      Apache 2.0
    • A
      license
      Not graded
      quality
      C
      maintenance
      Secure MCP protocol proxy with OAuth2 + Dynamic Client Registration (DCR), JWT auth, RBAC, rate limiting, multi-server aggregation, and a monitoring/admin dashboard.
      72 PyPI
      13
      MIT
    • A
      license
      A
      quality
      B
      maintenance
      MCP server that gives agents the same interface as dashAI's GUI: explore datasets, list models, train, track jobs, and read metrics.
      10
      36 PyPI
      MIT