home / mcp / finance mcp server

Finance MCP Server

Provides modular financial research flows via an MCP server with zero-code configuration and multi-protocol access.

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "flowllm-ai-finance-mcp": {
      "command": "uvx",
      "args": [
        "finance-mcp",
        "config=default,ths",
        "mcp.transport=stdio",
        "llm.default.model_name=qwen3-30b-a3b-thinking-2507",
        "disabled_flows='[\"tavily_search\",\"mock_search\",\"react_agent\"]'"
      ],
      "env": {
        "TAVILY_API_KEY": "xxx",
        "FLOW_LLM_API_KEY": "xxx",
        "DASHSCOPE_API_KEY": "xxx",
        "FLOW_LLM_BASE_URL": "https://dashscope.aliyuncs.com/compatible-mode/v1",
        "TUSHARE_API_TOKEN": "xxx",
        "BAILIAN_MCP_API_KEY": "xxx"
      }
    }
  }
}

Finance MCP is an intelligent agent toolkit and MCP server designed for financial research scenarios. Built on the FlowLLM framework, it integrates components such as Crawl4AI, Tushare, Tavily/DashScope search, and more to help you quickly build professional financial research agent systems while supporting zero-code configuration and multi-protocol access.

How to use

Connect to the Finance MCP server using an MCP client to run research workflows, perform web searches, extract entities from text, and orchestrate multiple tools in a single session. You can run in local stdio mode for development or start a remote HTTP/SSE server for remote access. Disable any flows you don’t need by adjusting configuration so you only run the tools you have keys for.

How to install

Prerequisites: you need Python and a runtime tool able to run MCP servers. Install the Finance MCP package using Python’s package manager.

pip install finance-mcp

Server configuration and usage

You can run the server in stdio mode for local development or in HTTP/SSE mode for remote usage. The stdio mode is configured to communicate via standard input/output and is suitable for connecting from local MCP clients.

{
  "mcpServers": {
    "finance_mcp": {
      "command": "uvx",
      "args": [
        "finance-mcp",
        "config=default,ths",
        "mcp.transport=stdio",
        "llm.default.model_name=qwen3-30b-a3b-thinking-2507",
        "disabled_flows='[\"tavily_search\",\"mock_search\",\"react_agent\"]'"
      ],
      "env": {
        "FLOW_LLM_API_KEY": "xxx",
        "FLOW_LLM_BASE_URL": "https://dashscope.aliyuncs.com/compatible-mode/v1",
        "DASHSCOPE_API_KEY": "xxx",
        "TUSHARE_API_TOKEN": "xxx",
        "TAVILY_API_KEY": "xxx",
        "BAILIAN_MCP_API_KEY": "xxx"
      }
    }
  }
}

Starting via HTTP RESTful API with streaming

To enable remote access over HTTP with streaming, start the server in HTTP backend mode. You then expose flows configured for streaming and send requests to HTTP endpoints that stream responses via SSE.

finance-mcp \
  config=default,stream_agent \
  backend=http \
  http.host=0.0.0.0 \
  http.port=8002 \
  llm.default.model_name=qwen3-30b-a3b-thinking-2507

Connecting from an MCP client

If you start the server with the HTTP backend, configure your MCP client to connect to the remote SSE server using the provided URL. You can also work with a FastMCP client to directly issue tool calls and receive streaming results.

{
  "mcpServers": {
    "finance_mcp": {
      "type": "sse",
      "url": "http://0.0.0.0:8001/sse"
    }
  }
}

Available tools

history_calculate

Price-volume analysis based on Tushare A-share historical data.

crawl_url

Scrape and parse web content.

extract_entities_code

Identify financial entities from text and complete stock codes (dashscope_search integration).

execute_code

Execute arbitrary Python code.

execute_shell

Execute shell commands.

dashscope_search

Web search based on DashScope.

tavily_search

Web search based on Tavily.

mock_search

Mock search for LLM simulation.

react_agent

ReAct agent combining multiple tools for complex queries.

tongyi_search

WebSearch service based on DashScope external MCP.

bochaai_search

BochaAI search service based on DashScope external MCP.