fred-mcp-server
An MCP server that provides access to the full FRED API (Federal Reserve Economic Data), including the Maps API for regional and geographic data.
README
fred-mcp-server
An MCP (Model Context Protocol) server that provides access to the full FRED API (Federal Reserve Economic Data), including the Maps API for regional and geographic data. Use it to search, explore, and retrieve economic data directly within Claude conversations.
Disclaimer: This project is not affiliated with, endorsed by, or connected to the Federal Reserve Bank of St. Louis or any Federal Reserve entity. It is an independent open-source tool that accesses the publicly available FRED API.
Features
- 33 tools covering all FRED API endpoints including Maps (series, categories, releases, sources, tags, maps)
- Full parameter support — no artificial limits on pagination or filtering
- Built-in rate limiting (120 requests/minute)
- Async HTTP client for efficient request handling
Requirements
- Python 3.10+
- A free FRED API key
Installation
pip install fred-mcp-server
Or with uv:
uv tool install fred-mcp-server
With Docker:
docker pull ghcr.io/floriancaro/fred-mcp-server
Or build from source:
docker build -t fred-mcp-server https://github.com/floriancaro/fred-mcp-server.git
From source:
git clone https://github.com/floriancaro/fred-mcp-server.git
cd fred-mcp-server
uv sync --dev
Configuration
After installing, configure the MCP server for your preferred client.
Claude Code (CLI)
Add the server globally (available in all projects):
claude mcp add fred -s user -e FRED_API_KEY=your-api-key-here -- fred-mcp-server
Or add it to a specific project only:
claude mcp add fred -e FRED_API_KEY=your-api-key-here -- fred-mcp-server
Verify it's connected:
claude mcp list
Claude Code (project config)
Alternatively, create a .mcp.json file in your project root:
{
"mcpServers": {
"fred": {
"command": "fred-mcp-server",
"env": {
"FRED_API_KEY": "your-api-key-here"
}
}
}
}
A .mcp.json.example file is included in the repo as a template.
Claude Desktop
Add to your Claude Desktop config file:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"fred": {
"command": "fred-mcp-server",
"env": {
"FRED_API_KEY": "your-api-key-here"
}
}
}
}
Restart Claude Desktop after saving.
Docker
If using the Docker image, replace the command field:
{
"mcpServers": {
"fred": {
"command": "docker",
"args": ["run", "-i", "--rm", "-e", "FRED_API_KEY", "ghcr.io/floriancaro/fred-mcp-server"],
"env": {
"FRED_API_KEY": "your-api-key-here"
}
}
}
}
Tools
Series
| Tool | Description |
|---|---|
fred_series |
Get metadata for a series |
fred_series_search |
Search for series by text |
fred_series_observations |
Get data values for a series |
fred_series_categories |
Get categories for a series |
fred_series_release |
Get the release for a series |
fred_series_tags |
Get tags for a series |
fred_series_search_tags |
Get tags matching a search query |
fred_series_search_related_tags |
Get related tags for a search query |
fred_series_updates |
Get recently updated series |
fred_series_vintagedates |
Get vintage dates for a series |
Categories
| Tool | Description |
|---|---|
fred_category |
Get a category (root = 0) |
fred_category_children |
Get child categories |
fred_category_related |
Get related categories |
fred_category_series |
Get series in a category |
fred_category_tags |
Get tags for a category, or related tags via related_to |
Releases
| Tool | Description |
|---|---|
fred_releases |
List all releases |
fred_releases_dates |
Get dates for all releases |
fred_release |
Get a specific release |
fred_release_dates |
Get dates for a release |
fred_release_series |
Get series in a release |
fred_release_sources |
Get sources for a release |
fred_release_tags |
Get tags for a release, or related tags via related_to |
fred_release_tables |
Get release table trees |
Sources
| Tool | Description |
|---|---|
fred_sources |
List all sources |
fred_source |
Get a specific source |
fred_source_releases |
Get releases for a source |
Tags
| Tool | Description |
|---|---|
fred_tags |
List/search all tags |
fred_related_tags |
Get related tags |
fred_tags_series |
Get series matching tags |
Maps
| Tool | Description |
|---|---|
geofred_series_group |
Get metadata for a geographic FRED series |
geofred_series_data |
Get cross-sectional regional data for a geographic series |
geofred_regional_data |
Get cross-sectional regional data by series group |
geofred_shapes |
Get GeoJSON shape files for geographic region boundaries |
Example Prompts
Once configured, you can ask Claude things like:
- "What is the current US GDP growth rate?" (uses
fred_series_search+fred_series_observations) - "Show me the unemployment rate for the past 10 years" (uses
fred_series_observationswithobservation_start) - "What data releases are coming up this week?" (uses
fred_releases_dates) - "Find all series related to housing starts" (uses
fred_series_search) - "Compare regional unemployment rates across states" (uses
geofred_regional_data)
Development
git clone https://github.com/floriancaro/fred-mcp-server.git
cd fred-mcp-server
uv sync --dev
# Run unit tests
uv run pytest tests/test_client.py tests/test_tools.py -v
# Run integration tests (requires FRED_API_KEY)
FRED_API_KEY=your-key uv run pytest tests/test_integration.py -v
Troubleshooting
"FRED_API_KEY environment variable is not set"
Ensure the FRED_API_KEY is passed in your MCP configuration's env block, or set it in your shell environment.
Rate limiting The server limits requests to 120 per minute (matching FRED API limits). If you hit the limit, requests will automatically wait — no action needed.
Verifying the server is running
claude mcp list
Contributing
Contributions are welcome! Please open an issue or submit a pull request.
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。