home / mcp / finance tools mcp server

Finance Tools MCP Server

A Model Context Protocol server for building an investor agent

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "voxlink-org-finance-tools-mcp": {
      "command": "uvx",
      "args": [
        "finance-tools-mcp"
      ],
      "env": {
        "FRED_API_KEY": "YOUR_API_KEY"
      }
    }
  }
}

Finance Tools MCP Server enables Large Language Models to perform in-depth financial analysis by connecting to data sources and analytical libraries. You can run it locally with an MCP-compatible client, supply data via environment variables, and use it to power financial research, macro data, and tooling within conversations.

How to use

You run Finance Tools MCP Server locally and connect to it from your MCP client. Start the server with the standard runtime tool, optionally providing a macroeconomic data API key for enhanced data access. You can choose a transport method that suits your client, such as a default HTTP-like stream or a Server-Sent Events (SSE) channel.

Key usage patterns: connect your MCP client to the running server, trigger financial analyses, fetch macro data when needed, and iterate on results within your workflows. If you enable the macro data API key, your analysis can pull richer macro context to support more accurate insights.

How to install

Prerequisites: ensure you have Python 3.10 or higher and a compatible MCP runtime tool installed on your system.

Install the MCP runtime tool (uv) if you have not already, then ensure it is available on your PATH.

Run the MCP server using the runtime tool with optional environment variables for enhanced data access.

# Step 1: Install uv (if not already installed)
curl -LsSf https://astral.sh/uv/install.sh | sh
# Make sure uv is on your PATH (may require a restart or PATH fix)
# Step 2: Run the MCP server
uvx finance-tools-mcp
# Optional: enable FRED macro data access
FRED_API_KEY=YOUR_API_KEY uvx finance-tools-mcp
# Optional: use SSE transport
uvx finance-tools-mcp --transport sse
# Combine both options
FRED_API_KEY=YOUR_API_KEY uvx finance-tools-mcp --transport sse

Additional sections

Configuration notes, debugging tips, and development guidance are provided below to help you integrate and troubleshoot the server in local and testing environments.

If you want to test integration with an MCP client or IDE, you can configure the client to point at the running server and, when needed, pass an API key for macroeconomic data access.

Debugging and logs

Use the MCP inspector to debug the server and verify that it is reachable by your MCP client.

npx @modelcontextprotocol/inspector uvx finance-tools-mcp

For log visibility, check log locations on your platform after starting the server.

Development notes

During development, you can test with a client configuration that points to the local runtime, using the directory where the server lives.

{
  "mcpServers": {
    "investor": {
      "command": "path/to/uv/command/uv",
      "args": ["--directory", "path/to/finance-tools-mcp", "run", "finance-tools-mcp"]
    }
  }
}