# financial-modeling-prep-mcp-server [Health: Active]

**Category:** 💰 Finance & Fintech  
**Repository:** https://github.com/imbenrabi/Financial-Modeling-Prep-MCP-Server  
**GitHub Stars:** 150  
**npm Downloads (last month):** 2460  
**Views:** 1  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/financial-modeling-prep-mcp-server

## Description
MCP server for Financial Modeling Prep API with 250+ financial data tools

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "financial-modeling-prep-mcp-server": {
    "command": "npx",
    "args": ["-y","financial-modeling-prep-mcp-server"],
    "env": {
      "FMP_ACCESS_TOKEN": ""
    }
  }
}
```

**Requires environment variables:** `FMP_ACCESS_TOKEN` — the values above are empty placeholders; fill in real credentials before running (see the repository for what each one is for).

## Documentation

## What financial-modeling-prep-mcp-server does

The financial-modeling-prep-mcp-server MCP server exposes Financial Modeling Prep data to MCP-compatible AI assistants. Its catalog contains more than 250 tools organized across 24 categories, including company information, financial statements, market quotes, technical indicators, SEC filings, earnings, news, ETFs, funds, cryptocurrency, forex, commodities, and economic data.

The server is intended for assistants that need to retrieve financial information through the Financial Modeling Prep API. It does not provide a separate financial database; requests are made through the connected Financial Modeling Prep service and require an API credential.

## How it works

Connections use HTTP/SSE transport. The hosted endpoint is available at `https://financial-modeling-prep-mcp-server-production.up.railway.app/mcp`, so a client can connect without installing the package locally. Self-hosted deployments can run through npm, Docker, or a source checkout.

Tool availability is controlled by the selected operating mode:

- Dynamic mode starts with five meta-tools and lets the assistant load toolsets during a session.
- Static mode preloads the toolsets named in configuration.
- All Tools mode, the default, loads the full catalog when the session starts.

The project uses toolception for runtime tool management in dynamic mode. CLI arguments take precedence over environment variables, followed by session configuration and then defaults.

## Setup and configuration

For a hosted connection, provide `FMP_ACCESS_TOKEN` in the session configuration. For a local npm deployment, run `npx financial-modeling-prep-mcp-server --fmp-token=YOUR_FMP_API_KEY`, or install the package globally and start it with `fmp-mcp --fmp-token=YOUR_FMP_API_KEY`.

The documented prerequisite is Node.js 25.3.0 or newer for version 2.6.0 and later. Versions 2.5.x and earlier require Node.js 20 or newer. Docker images can be built from source or pulled as pre-built images. Source installation uses `npm install` followed by `npm run build`.

Dynamic mode is enabled with `DYNAMIC_TOOL_DISCOVERY=true`. Static mode uses `FMP_TOOL_SETS`, such as `search,company`, to select toolsets before session creation. Omitting both settings selects the all-tools default.

## Tools and capabilities

The financial-modeling-prep-mcp-server MCP server groups its tools into areas such as:

- Search, directories, and symbol lists
- Company profiles, financial statements, metrics, and analysis
- Quotes, prices, market data, indexes, and performance
- Technical indicators and earnings
- News, press releases, SEC filings, and trading activity
- ETFs, funds, government trading, crypto, forex, commodities, and economics
- Bulk data and other specialized datasets

This organization lets a deployment expose the complete catalog or limit the initial session to selected toolsets. Dynamic discovery is useful when an assistant should add capabilities only when needed, while static configuration is suitable when the required groups are known in advance.

## Limitations and notes

An API key from Financial Modeling Prep is required. The README points users to Financial Modeling Prep to obtain one, but it does not specify plan limits, endpoint coverage by subscription, or data licensing terms.

Loading all tools is the default behavior, which may be less suitable for clients or sessions where a smaller tool list is preferred. Dynamic and static modes provide alternatives, but the available toolsets still depend on the Financial Modeling Prep API.

The documented client guidance covers Claude.ai and Claude Desktop as remote HTTP/SSE connections, plus custom clients that can POST to `/mcp` with an `mcp-client-id` header. The project is also listed in Smithery.ai and Glama.ai registries.

_Full upstream README: https://allmcps.com/mcp/financial-modeling-prep-mcp-server/readme_