@daften/fireflyiii-mcp

@daften/fireflyiii-mcp

Connects any MCP-compatible AI assistant to your Firefly III personal finance instance, enabling natural language queries about your finances.

Category
访问服务器

README

MCP server for Firefly III

npm version npm downloads CI License: MIT Documentation

An MCP (Model Context Protocol) server that connects any MCP-compatible AI assistant to your Firefly III personal finance instance. Ask your AI assistant questions about your finances in natural language.

📖 Full documentation → daften.github.io/fireflyiii-mcp

What you can ask

Once configured, you can ask things like:

  • "How much did I spend on groceries last month?"
  • "Show me my budget status for this month."
  • "Find any duplicate transactions in the last 30 days."
  • "Set up a piggy bank for my vacation fund with a €2000 target."
  • "What were my biggest expense categories this year?"

Your AI assistant handles the Firefly III API calls — you get answers in plain language.


Choose your setup method:

Method Transport Best for
npm — stdio stdio Simplest setup, AI on the same machine
npm — HTTP HTTP + OAuth or PAT Remote AI access, or a headless gateway with no browser in the loop
Docker — HTTP HTTP + OAuth or PAT Self-hosted on a server or home lab
Git checkout stdio or HTTP Contributing or local development

All options except Docker require Node.js 20+.


Option 1: npm package — stdio (simplest)

Requires: Node.js 20+, a Firefly III Personal Access Token (Options → Remote access and tokens → Create new token).

Add to your Claude MCP config (.claude/mcp.json or Claude Desktop claude_desktop_config.json):

{
  "mcpServers": {
    "fireflyiii": {
      "command": "npx",
      "args": ["-y", "@daften/fireflyiii-mcp"],
      "env": {
        "FIREFLY_URL": "https://your-firefly-instance.example.com",
        "FIREFLY_TOKEN": "your-personal-access-token-here"
      }
    }
  }
}

Your MCP client downloads and starts the server automatically on first use. No separate install step needed.


Option 2: npm package — HTTP (OAuth or PAT)

→ See the HTTP/OAuth setup guide, or HTTP/PAT for headless callers (gateways, automation) that can't drive a browser-based OAuth flow.


Option 3: Docker — HTTP (self-hosted)

→ See Docker setup guide in the docs.


Option 4: Git checkout (development)

→ See Git checkout guide in the docs.


Nightly builds (unstable)

Want to test unreleased changes from main? A nightly build is published automatically each night that main has changed. These are unstable and not recommended for production.

  • npm: npm install @daften/fireflyiii-mcp@nightly (or npx @daften/fireflyiii-mcp@nightly)
  • Docker: docker pull ghcr.io/daften/fireflyiii-mcp:nightly

A normal install (no tag) always resolves to the latest tagged release — @latest on npm and :latest on Docker are never moved to a nightly. To go back to a stable build, reinstall without the @nightly / :nightly tag.


Experimental Autocomplete Prompts

→ See Autocomplete prompts in the docs.


Available Tools

→ See the full tool reference in the docs (140 tools across 14 groups).


Filtering Tools

→ See Tool filtering in the docs.


Development

npm test                  # Run unit tests
npm run test:watch        # Watch mode
npm run test:integration  # Run against live Firefly III (requires FIREFLY_URL + FIREFLY_TOKEN)
npm run dev               # Run without building (uses tsx)
npm run build             # Compile TypeScript to dist/

Resources

Contributing

See CONTRIBUTING.md for the development loop, tool-add checklist, and commit conventions.

Security

See SECURITY.md for the vulnerability disclosure policy.

Acknowledgements

Feature comparison informed by fabianonetto/mcp-server-firefly-iii and etnperlong/firefly-iii-mcp.

License

MIT

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选