MCP.so
Sign In
P

penDART MCP - Korean corporate disclosure data for AI agents

@SongHyojun0228

Repository

About penDART MCP - Korean corporate disclosure data for AI agents

### Server Name

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "opendart": {
      "command": "npx",
      "args": [
        "-y",
        "opendart-mcp"
      ],
      "env": {
        "DART_API_KEY": "<YOUR_API_KEY>"
      }
    }
  }
}

Tools

5

Search for a Korean company and get basic corporate info (한국 기업 검색 및 기본 정보 조회). Input can be a company name (삼성전자), stock code (005930), or DART corp code (00126380).

Search DART disclosure filings (공시 보고서 검색). Can search by company or browse all recent filings. Returns filing title, date, and DART link.

Get key financial data for a Korean company (기업 주요 재무 데이터 조회). Returns revenue, operating profit, net income, total assets/liabilities/equity with year-over-year change rates.

Compare financial data across multiple Korean companies (다중 회사 재무 비교). Shows revenue, operating profit, and net income side by side.

Get full financial statements with all account items (전체 재무제표 상세 조회). Returns every line item of BS, IS, CIS, CF, or SCE.

Overview

What is penDART MCP - Korean corporate disclosure data for AI agents?

penDART MCP connects AI agents (Claude Desktop, Cursor, etc.) to Korea’s DART (Electronic Disclosure System) official corporate filings. It lets you ask natural‑language questions like “Show me Samsung Electronics’ financial statements” and get real answers from public Korean financial data.

How to use penDART MCP - Korean corporate disclosure data for AI agents?

Add the server to your AI client’s MCP configuration (e.g., claude_desktop_config.json). Run with npx opendart-mcp and set the DART_API_KEY environment variable. A free API key is obtained from opendart.fss.or.kr. Corporate codes are downloaded automatically on first start.

Key features of penDART MCP - Korean corporate disclosure data for AI agents

  • Search Korean companies by name, stock code, or DART corp code.
  • Search disclosure filings with date/type filters.
  • Get key financials: revenue, operating profit, net income, assets, liabilities.
  • Compare financials across multiple companies side by side.
  • Retrieve full financial statements (balance sheet, income statement, cash flow, etc.).

Use cases of penDART MCP - Korean corporate disclosure data for AI agents

  • Ask your AI for a company’s basic information and stock code.
  • Query recent disclosure filings for a specific company.
  • Request annual or quarterly financial summaries (e.g., revenue, net income).
  • Compare financial performance of competing Korean companies.
  • Retrieve detailed financial statements for in‑house analysis.

FAQ from penDART MCP - Korean corporate disclosure data for AI agents

How do I get a DART API key?

Sign up at opendart.fss.or.kr, log in, and apply for an API key. It is issued instantly and free of charge.

Is the API key free? Are there usage limits?

Yes, it is free. The DART OpenAPI allows up to 10,000 requests per day.

Do I need to manually download the corporate code data?

No. The server automatically downloads the corporate code database on first run. You can manually update it later with npm run update-corp-codes if needed.

Which AI clients are supported?

Claude Desktop and Cursor are explicitly supported. Any MCP‑compatible client (including MCP Inspector for debugging) can be used.

What is the project structure?

The server entry point is src/index.ts. Tools are organized in src/tools/ (disclosure and financial) and utilities in src/utils/ (DART API client, corp‑code conversion, formatting helpers).

Frequently asked questions

How do I get a DART API key?

Sign up at [opendart.fss.or.kr](https://opendart.fss.or.kr), log in, and apply for an API key. It is issued instantly and free of charge.

Is the API key free? Are there usage limits?

Yes, it is free. The DART OpenAPI allows up to 10,000 requests per day.

Do I need to manually download the corporate code data?

No. The server automatically downloads the corporate code database on first run. You can manually update it later with `npm run update-corp-codes` if needed.

Which AI clients are supported?

Claude Desktop and Cursor are explicitly supported. Any MCP‑compatible client (including MCP Inspector for debugging) can be used.

What is the project structure?

The server entry point is `src/index.ts`. Tools are organized in `src/tools/` (disclosure and financial) and utilities in `src/utils/` (DART API client, corp‑code conversion, formatting helpers).

Comments

Basic information

Category

Data & Analytics

Transports

stdio

Author

SongHyojun0228

Submitted by

효준 송

More Data & Analytics MCP servers

B

BidSkim - UK Tenders & Contract Renewals

BidSkim

UK procurement intelligence for AI assistants: search live UK government tenders, find contract renewals with expiry dates and incumbent suppliers, and profile public-sector buyers and suppliers. Covers Find a Tender, Co

Batru

Batru

Real, calibrated win-rate predictions, pick recommendations, counters, synergy and tier lists for Dota 2, Deadlock and Marvel Rivals, from production models trained on ~20M matches. Free hosted endpoint (Streamable HTTP)

Subtext

Subtext by Fullstory

Session replay, built for agents. Subtext is agentic session review: it captures production sessions of your app and connects them to your coding agent — Claude Code, Cursor, Codex, Devin, your own harness — so it can

9

TofuBofu AI Visibility

Arnav Neil

TofuBofu runs a free AI visibility scan for any B2B company, right inside your AI assistant. Ask it to scan a domain and it runs the questions real buyers ask across ChatGPT, Claude, Gemini, Perplexity, Google AI Overvie

Minds: Synthetic Market Research Panels

Minds

Minds is a synthetic market research platform. This MCP server lets ChatGPT, Claude, and Cursor run customer research end to end without leaving the assistant. Describe an audience in a brief ("German Gen Z grocery shopp