ragalgo

@kokogo100/ragalgo

0 Stars 1 次浏览 kokogo100 更新于 2026-08-23

Dynamic RAG Engine preventing AI hallucinations in Korean Finance and Crypto markets.

金融

MCP 服务配置

复制以下 JSON 到 OPClaw 或其他 MCP 客户端的配置文件中即可使用

{
  "mcpServers": {
    "ragalgo-mcp-server": {
      "args": [
        "ragalgo-mcp-server@1.0.4"
      ],
      "command": "npx"
    }
  }
}

可用工具（5 个）

该服务在 MCP 协议中暴露的工具，AI 可按需调用

tavily_search 14 个参数需填 1 项 Search the web for current information on any topic. Use for news, facts, or data beyond your knowledge cutoff. Returns snippets and source URLs.

Search the web for current information on any topic. Use for news, facts, or data beyond your knowledge cutoff. Returns snippets and source URLs.

必填参数：query

tavily_extract 6 个参数需填 1 项 Extract content from URLs. Returns raw page content in markdown or text format.

Extract content from URLs. Returns raw page content in markdown or text format.

必填参数：urls

tavily_crawl 11 个参数需填 1 项 Crawl a website starting from a URL. Extracts content from pages with configurable depth and breadth.

Crawl a website starting from a URL. Extracts content from pages with configurable depth and breadth.

必填参数：url

tavily_map 8 个参数需填 1 项 Map a website's structure. Returns a list of URLs found starting from the base URL.

Map a website's structure. Returns a list of URLs found starting from the base URL.

必填参数：url

tavily_research 2 个参数需填 1 项 Perform comprehensive research on a given topic or question. Use this tool when you need to gather information from multiple sources to answer a question or complete a task. Returns a detailed response based on the research findings.

Perform comprehensive research on a given topic or question. Use this tool when you need to gather information from multiple sources to answer a question or complete a task. Returns a detailed response based on the research findings.

必填参数：input

服务介绍

RagAlgo: Dynamic RAG Engine for AI Reliability

"Your AI is an Analyst, NOT a Day Trader."

RagAlgo is an MCP Server that provides mathematically scored financial context (Korean Stocks/Crypto) to AI agents.
We focus on "State-of-Truth" (Daily Closed Data) to prevent AI hallucinations caused by real-time market noise.

Analyst, Not Broker: We provide "Daily Analysis Reports" (Post-Market), not real-time tick data.
Scored Context: Instead of raw prices, we give you "Scores" (0~100) and "Zones" (Forest vs Tree).
Global Market Specialist: Optimized for US, UK, JP, KR, and Crypto.

👉 Official Website (ragalgo.com)

# 📖 Architecture & Whitepaper

Discover why RagAlgo is the "Hippocampus" for Agentic AI, not just another RAG.

# # 🏗️ Data Pipeline Architecture

Our production system on Railway processes global financial data 24/7:

┌─────────────────────────────────────────────────────────────────────────────┐
│                        RagAlgo Data Pipeline (Railway)                      │
├─────────────────────────────────────────────────────────────────────────────┤
│                                                                             │
│  📥 COLLECT              🔍 FILTER              🏷️ TAG           📊 SCORE  │
│  ─────────────          ──────────────         ─────────        ──────────  │
│  • KR-News-Collector    • filter-worker-1      • tag-worker     • Gemini-1  │
│  • US-News-Collector    • filter-worker-2      • Meta-Hierarchy • Gemini-2  │
│  • UK-News-Collector    • filter-worker-3      •   Worker       • ...       │
│  • JP-News-Collector    • ibkr_filter_worker3  │                • Gemini-7  │
│  • research-collector   │                      │                │           │
│                         │                      │                │           │
│  ════════════════════════════════════════════════════════════════════════  │
│                                    ↓                                        │
│                      📦 SNAPSHOT (Daily 18:00 KST)                          │
│                      ──────────────────────────────                         │
│                      • KR-Snapshot  • US-Snapshot                           │
│                      • UK-Snapshot  • JP-Snapshot                           │
│                      • CRY-Snapshot • Unified-Snapshot                      │
│                                    ↓                                        │
│                          🚀 SERVE (MCP Server)                              │
│                          ─────────────────────                              │
│                          • RagAlgo-Service (SSE/stdio)                      │
│                          • ragalgo-relay-server (WebSocket)                 │
│                                                                             │
└─────────────────────────────────────────────────────────────────────────────┘

RagAlgo Architecture

Vision Whitepaper (The Hook)
- Concept: Why RagAlgo is a "Semantic Digital Twin" (SDT) using the Hippocampus analogy.
- Value: Explains the "Self-Growing Taxonomy" and "Data Flywheel" effect.
Technical Report (The Proof)
- Deep Dive: Detailed anatomy of the Contextual Knowledge Network (CKN).
- 2025 Trend: How RagAlgo serves as the memory layer for Agentic AI (e.g., PepsiCo/Salesforce Agentforce).

# 💡 Why "Daily Close"?

Users often ask: "Why isn't the chart data real-time?"

Because AI performs better with clarity.
Real-time tick data is full of noise and volatility. If you feed an LLM raw live prices, it often hallucinates patterns that don't exist.

RagAlgo acts like a Professional Technical Analyst who works after the market closes:

Wait for the dust to settle (Market Close).
Analyze the day's battle (Daily Candle & Aux Indicators).
Deliver a "Confirmed Strategy" to your AI.

Use RagAlgo to build "Investment Advisors", not "High-Frequency Trading Bots".

# 🚀 Quick Start

# # Claude Desktop Configuration

Add this to your config file:

Windows: %APPDATA%\Claude\claude_desktop_config.json
Mac: ~/Library/Application Support/Claude/claude_desktop_config.json

# # # ☁️ Cloud Mode (Recommended - No installation required)

{
  "mcpServers": {
    "ragalgo": {
      "url": "https://ragalgo-service-production.up.railway.app/sse",
      "env": {
        "RAGALGO_API_KEY": "YOUR_API_KEY_HERE"
      }
    }
  }
}

# # # 📦 Local Mode (Requires Node.js)

{
  "mcpServers": {
    "ragalgo": {
      "command": "npx",
      "args": ["-y", "ragalgo-mcp-server", "- -stdio"],
      "env": {
        "RAGALGO_API_KEY": "YOUR_API_KEY_HERE"
      }
    }
  }
}

Tip: You can get a Free 1,000 Call Key instantly at RagAlgo Dashboard.

# 📚 Usage Examples (Cookbook)

We have a dedicated repository for practical examples to help you get started quickly.
Please visit the RagAlgo Examples Repository.

# # What's Inside?

8 Step-by-Step Recipes: From basic data fetching to advanced AI agents.
Skeleton Code + Prompts: Copy-paste ready resources.
Scenarios:
- 🐣 Basic: Get stock scores in 5 minutes.
- 🧪 Intermediate: Verify technical signals with AI.
- 🚀 Advanced: Build an autonomous reasoning agent (Mock Trading Audit).
- ☕ Morning Briefing: Create a bot that emails you a daily market summary.

"Skeleton + Prompt" Approach: We provide the ingredients. You ask ChatGPT/Claude to cook!

# 🌍 Supported Markets & Roadmap

RagAlgo is expanding its CKN coverage globally. Currently, US, UK, Japan, Korea, and Crypto markets are fully supported.

# 🛠 Tools

⚠️ CORE CONCEPT: Scored vs Raw

get_news_scored (Default): Returns only significant news (Scores ≠ 0). Best for AI decision making.

get_news (Raw): Returns ALL news including noise. Use this ONLY if you need raw data feed.

# 📡 Real-time WebSocket (Business Tier)

For users who really need live data (e.g., for monitoring dashboards), we offer a WebSocket stream.
Note: This is strictly for monitoring, not for LLM inference context.

Access: Business Plan subscribers (Includes 30 connections).
Address: wss://ragalgo-relay-server-1-production.up.railway.app
Guide: See Developer Docs for implementation details.

# 💬 Support

Website: ragalgo.com
Email: support@ragalgo.com

查看源码项目主页