fireflyiii-mcp

Connect any MCP-compatible AI assistant to your Firefly III personal finance instance. Query your finances in natural language — no queries, no code.

Get Started

GitHub

Docker (GHCR)

🛠️

140 tools

14 groups covering accounts, transactions, budgets, categories, bills, piggy banks, reports, rules, recurring transactions, attachments, currencies, exports, object groups, and transaction links.

🔌

stdio & HTTP

Connect locally via Personal Access Token (stdio), or self-host with OAuth 2.0 for remote access from any MCP-compatible client.

🎛️

Tool filtering

Load only what you need. Choose a named preset or pick individual groups. Add --read-only to restrict to safe, non-destructive tools.

🐳

Docker ready

Run as a self-hosted container. Pull from GHCR and connect any MCP-compatible AI assistant remotely over HTTP.

What you can ask

Once configured, ask your AI assistant things like:

"How much did I spend on groceries last month?"

"Find any duplicate transactions in the last 30 days."

"Set up a piggy bank for my vacation fund with a €2000 target, linked to my savings account."

"Which bills are coming up this week and am I on track?"

"What were my biggest expense categories this year?"

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

Quickstart

The simplest setup uses npx with stdio transport and a Personal Access Token (PAT).

Step 1: Create a PAT in Firefly III: Options → Remote access and tokens → Create new token

Step 2: Add to your MCP client 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 AI assistant downloads and starts the server automatically on first use. → Full stdio guide

All setup options

Method	Transport	Best for
npm — stdio	stdio	Simplest setup, AI on the same machine
npm — HTTP/OAuth	HTTP + OAuth 2.0	Remote AI access, OAuth instead of PAT
Docker — HTTP	HTTP + OAuth 2.0	Self-hosted on a server or home lab
Git checkout	stdio or HTTP	Contributing or local development