RagAlgo: Dynamic RAG Engine for AI Reliability

Global Beta: Currently specialized in Korean Finance (Stocks & Crypto), but built on a universal "Tag & Score" protocol applicable to any domain.

RagAlgo is not just a data API. It is a "State-of-Truth" Provider designed to prevent LLM hallucinations when dealing with complex static knowledge and highly volatile dynamic data.

Why RagAlgo?

Traditional RAG (SQL/Vector DB) often fails to capture the "current context" of fast-changing reality. RagAlgo solves this by converting data into Mathematical Scores and Standardized Tags.

Core Philosophy

1. Sanitized Data: We filter out noise so AI only ingests high-quality signals.
2. Mathematical Scoring: We translate complex movements into simple scores (e.g., Chart Score: 9.5), giving AI a clear ground truth to reason upon.
3. Universal Protocol: Our "Tag & Score" architecture is designed to systematize and standardize any data domainincluding Legal, Medical, and Techinto a format that AI can flawlessly understand and judge.

Flexible Integration: Scored vs. Non-Scored

We respect your freedom to build. Choose the mode that fits your pipeline:

RagAlgo MCP Server (KOR)

API MCP(Model Context Protocol)  .

cd mcp-server
npm install
npm run build

# Windows (PowerShell)
$env:RAGALGO_API_KEY="ragalgo_scored_test_a3ed9bd570436d46"

# Windows (CMD)
set RAGALGO_API_KEY=ragalgo_scored_test_a3ed9bd570436d46

# Mac/Linux
export RAGALGO_API_KEY="ragalgo_scored_test_a3ed9bd570436d46"

Claude Desktop

%APPDATA%\Claude\claude_desktop_config.json :

{
  "mcpServers": {
    "ragalgo": {
      "command": "node",
      "args": ["C:\\coding\\RagAlgo\\mcp-server\\dist\\index.js"],
      "env": {
        "RAGALGO_API_KEY": "ragalgo_scored_test_a3ed9bd570436d46"
      }
    }
  }
}

(9)

Claude Desktop

"   5 "
 get_news_scored(tag="", limit=5)

"    "
 get_news_scored(tag="", verdict="bullish")

"005930  "
 get_financials(ticker="005930")

"  "
 get_chart_coin(ticker="KRW-BTC")

API

#   
npm run dev

# 
npm run build

#  
npm start