Quant Research Platform MCP Server

Quant Research Platform MCP Server

Enables AI assistants to directly access quant research knowledge, including factor libraries, strategy backtesting, and research reports, through the MCP protocol.

Category
Visit Server

README

Quant Research Platform

一个 LLM 原生的量化研究平台,提供因子管理、策略回测、研报分析等功能,并通过 MCP 协议让 AI 助手直接访问量化知识库。

Features

  • 因子知识库 - 因子管理、评分、分析
  • 策略回测 - 策略管理、参数搜索
  • 研报知识库 - PDF 上传、RAG 对话
  • 经验知识库 - 结构化研究经验、语义检索、知识提炼
  • 数据服务 - K线数据、因子计算
  • MCP 协议 - LLM 直接访问量化知识库

Quick Start

1. 安装依赖

按顺序安装以下依赖:

序号 依赖 macOS / Linux Windows
1 Docker 下载 Docker Desktop 下载 Docker Desktop
2 Node.js 20+ 下载brew install node 下载安装包
3 pnpm npm install -g pnpm npm install -g pnpm
4 uv curl -LsSf https://astral.sh/uv/install.sh | sh irm https://astral.sh/uv/install.ps1 | iex

Windows 提示: 安装完成后需要重启终端使环境变量生效

2. 配置环境变量

# 复制环境变量模板
cp .env.example .env

编辑 .env 文件,填入以下必要配置:

配置项 说明 示例
PRE_DATA_PATH 预处理K线数据路径 /Users/xxx/Downloads/coin-binance-spot-swap-preprocess-pkl-1h
COIN_CAP_PATH 市值数据路径 (可选) /Users/xxx/Downloads/coin-cap
LLM_API_URL LLM API 地址 https://api.openai.com/v1
LLM_API_KEY LLM API 密钥 sk-xxx

3. 启动项目

macOS / Linux:

# 克隆项目
git clone https://github.com/your-username/quant-research-platform.git
cd quant-research-platform

# 启动(自动安装依赖、启动数据库、启动服务)
make start local

# 查看服务状态
make status

# 查看日志
make logs local

# 停止服务
make stop local

Windows (PowerShell):

# 一键启动(自动安装依赖、启动所有服务、打开浏览器)
uv run python scripts/dev.py start

# 停止所有服务
uv run python scripts/dev.py stop

# 查看服务状态
uv run python scripts/dev.py status

# 查看日志
uv run python scripts/dev.py logs              # 全部服务
uv run python scripts/dev.py logs api           # 仅 API
uv run python scripts/dev.py logs -n 100 mcp-factor  # 最近 100 行

scripts/dev.py 是跨平台任务运行器,也可在 macOS/Linux 上使用,等价于 make start local。 该脚本会自动启动 PostgreSQL、Redis、API、7 个 MCP 服务和前端,无需手动开多个终端。

4. 访问服务

启动后访问:

  • 前端界面: http://127.0.0.1:5173
  • API 文档: http://127.0.0.1:8000/docs

MCP Integration

通过 MCP 协议,LLM 可以直接访问平台的所有功能。

Claude 配置

.mcp.json 中添加:

{
  "mcpServers": {
    "factor-hub": {
      "url": "http://localhost:6789/mcp"
    },
    "data-hub": {
      "url": "http://localhost:6790/mcp"
    },
    "strategy-hub": {
      "url": "http://localhost:6791/mcp"
    },
    "note-hub": {
      "url": "http://localhost:6792/mcp"
    },
    "research-hub": {
      "url": "http://localhost:6793/mcp"
    },
    "experience-hub": {
      "url": "http://localhost:6794/mcp"
    }
  }
}

Development

目录结构

quant-research-platform/
├── backend/
│   ├── app/routes/          # HTTP 路由层
│   ├── app/schemas/         # Pydantic 模型
│   └── domains/             # 业务域
│       ├── mcp_core/        # MCP 基础设施
│       ├── factor_hub/      # 因子知识库
│       ├── strategy_hub/    # 策略回测
│       ├── research_hub/    # 研报 RAG
│       ├── experience_hub/  # 经验知识库
│       ├── data_hub/        # 数据服务
│       └── note_hub/        # 笔记管理
├── frontend/
│   └── src/
│       ├── features/        # 功能模块
│       ├── components/      # 通用组件
│       └── lib/             # 工具库
├── docker/                  # 容器配置
├── docs/                    # 文档
└── factors/                 # 因子定义库

常用命令

# 开发
make start local          # 本地开发模式
make start dev            # Docker 开发模式
make logs-api             # API 实时日志
make logs-frontend        # 前端实时日志

# 生产
make start                # 生产模式
make healthcheck prod     # 健康检查

运行模式

模式 命令 说明
local make start local 本地运行 Python/Node,仅 Docker 运行数据库
dev make start dev 全容器化,支持热重载
prod make start 全容器化,优化构建

Tech Stack

Backend

  • FastAPI, SQLAlchemy 2.0, PostgreSQL 16 + pgvector
  • LangChain, Redis, aiohttp
  • structlog, ruff, pytest

Frontend

  • React 18, TypeScript, Vite 6
  • TanStack Query, Zustand, shadcn/ui
  • AG Grid, ECharts, Tailwind CSS

Infrastructure

  • Docker, Docker Compose, Caddy, Supervisor

License

MIT


重要更新

因子和截面因子文件已迁移至 private/ 目录。

# 旧位置(已废弃)
factors/
sections/

# 新位置
private/
  factors/      # 因子代码 (.py)
  sections/     # 截面因子 (.py)
  metadata/     # 因子元数据 (YAML)

private/ 目录已被 gitignore,需通过独立的私有仓库管理。详见 private/.gitkeep

Recommended Servers

playwright-mcp

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.

Official
Featured
TypeScript
Magic Component Platform (MCP)

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.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

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.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

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.

Official
Featured
TypeScript
Kagi MCP Server

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.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

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.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured