Richgo Finance MCP
Enables searching and analyzing Korean stocks using Richgo Finance data, including stock details, rankings, undervaluation, AI rankings, and curation screens.
README
Richgo Finance MCP
Model Context Protocol (MCP) server for public data exposed by
finance.richgo.ai.
It lets MCP clients search Korean stocks, fetch Richgo stock detail data, and use Richgo's ranking, undervaluation, AI ranking, and curation screens from an assistant.
This project only calls public Richgo Finance endpoints by default. It does not bypass authentication or scrape private account data.
Tools
richgo_search_stock: search by Korean stock name or ticker.richgo_get_stock_detail: fetch public detail for a ticker.richgo_analyze_stock: search/alias resolve, then fetch compact analysis.richgo_get_scores: score ranking table.richgo_get_undervalued: undervalued stock tabs.richgo_get_ai_rankings: AI ranking tab.richgo_get_curation: curated screener buckets.richgo_get_breakout: breakout/special signal endpoint.richgo_get_market_api: generic/api/market/...wrapper.richgo_get_public_api: constrained generic/api/...wrapper.richgo_get_consensus_picks: merge Richgo scores, undervaluation, gap, and AI rankings into one ranked candidate list.richgo_build_portfolio: build an integer-share portfolio from Richgo consensus signals for a budget and risk profile.richgo_save_obsidian_report: save a generated portfolio report to a user-provided Obsidian vault path.richgo_guided_portfolio: beginner-friendly portfolio wizard that uses MCP form elicitation when the client supports it.richgo_service_catalog: list exposed services.
Install
git clone https://github.com/reallygood83/rfmcp.git
cd rfmcp
npm install
npm run build
Run Locally
npm start
On the original author's Mac, Homebrew Node was broken, so this bundled Node path was used:
export PATH=/Users/moon/.cache/codex-runtimes/codex-primary-runtime/dependencies/node/bin:$PATH
npm run build
npm start
Codex or Claude MCP config
{
"mcpServers": {
"richgo-finance": {
"command": "node",
"args": ["/absolute/path/to/rfmcp/dist/index.js"]
}
}
}
Example Prompts
You do not need to know the MCP tool names. Ask naturally:
아이티센글로벌 리치고 데이터로 분석해줘/rf 셀트리온 분석해줘삼성전자와 SK하이닉스 리치고 상세 지표 비교해줘리치고 저평가 total 상위 10개 보여줘AI 랭킹 상위 종목을 요약해줘리치고 컨센서스 기준으로 오늘 투자 매력 높은 종목 10개 뽑아줘500만원 중립형 포트폴리오 만들어줘초보자용 리치고 포트폴리오 마법사 시작해줘나는 잘 모르니까 질문하면서 300만원 투자 전략 만들어줘오늘 리치고 기준으로 뭐가 제일 좋아?1000만원이면 몇 종목을 몇 주씩 사면 좋을까?이 결과를 내 Obsidian 볼트에 보고서로 저장해줘
Natural Language Routing
Most MCP clients choose tools from the descriptions below. These are the intended natural-language routes:
| User says | Preferred tool | What happens |
|---|---|---|
삼성전자 분석해줘, 005930 리치고로 봐줘 |
richgo_analyze_stock |
Resolves the stock and returns compact Richgo detail. |
저평가 상위 10개, AI 랭킹 보여줘 |
richgo_get_undervalued / richgo_get_ai_rankings |
Returns the requested screen. |
오늘 투자 매력 높은 종목, 뭐가 좋아?, 컨센서스 픽 |
richgo_get_consensus_picks |
Combines scores, undervaluation, gap, and AI ranking into one candidate list. |
500만원 포트폴리오, 몇 주씩 살까?, 예산으로 전략 짜줘 |
richgo_build_portfolio |
Builds an integer-share portfolio with cash balance and risks. |
초보자용, 질문하면서 도와줘, 마법사 시작 |
richgo_guided_portfolio |
Uses form elicitation when supported, then builds the portfolio. |
Obsidian에 저장, 볼트에 보고서로 저장 |
richgo_save_obsidian_report |
Saves only when a vault path is provided or configured privately. |
Optional Slash Command
This repo includes a Claude-style slash command template at commands/rf.md.
Install it as /rf with:
mkdir -p ~/.claude/commands
cp commands/rf.md ~/.claude/commands/rf.md
After that, use prompts like:
/rf 셀트리온 분석해줘
The /rf command is tuned to produce data-backed Korean reports, not just a
metric dump. It should connect revenue, profitability, cash flow, financial
safety, valuation, and Richgo percentile signals into a concrete story that is
easy to understand.
Portfolio and Obsidian reports
The portfolio tools intentionally do not hardcode a personal vault path. This keeps the published MCP reusable for other users.
Recommended natural-language flow:
초보자용 리치고 포트폴리오 마법사 시작해줘
If your MCP client supports form elicitation, the server asks for budget, risk profile, number of stocks, and whether to save to Obsidian. If the client does not support elicitation, call:
500만원 중립형 포트폴리오 만들어줘
To save Obsidian reports, either pass vaultPath in the tool call:
{
"budget": 5000000,
"riskProfile": "balanced",
"vaultPath": "/absolute/path/to/your/vault",
"noteDir": "Richgo"
}
Or keep the path in your private shell/MCP environment:
export RICHGO_OBSIDIAN_VAULT_PATH='/absolute/path/to/your/vault'
Do not commit personal vault paths to this repository.
Optional auth
Public endpoints work without auth. If you legitimately need login-only Richgo features, set your own cookie:
export RICHGO_COOKIE='...'
This server does not bypass authentication.
Disclaimer
Richgo Finance data is informational only and is not investment advice. Final investment decisions and responsibility belong to the user.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.