Trackor

Custom MCP server for tracking expenses with add/list/update/delete/summarize/export capabilities.

python

2

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "at0mxploit-trackor": {
      "command": "uv",
      "args": [
        "run",
        "fastmcp",
        "dev",
        ".\\\\main.py"
      ]
    }
  }
}

Trackor is a custom MCP server built with FastMCP and Streamlit that helps you manage expenses. It exposes a set of actions to add, view, summarize, update, delete, and export expenses, all accessible via an MCP client. This enables you to build workflows and dashboards around your expense data without writing separate API code.

How to use

Connect your MCP client to the Trackor server to perform common expense tasks. You can add new expenses, list or filter existing ones, update entries, delete by ID or date range, and export data in JSON or CSV formats. Use the provided tools through your MCP client’s interface to run the corresponding actions and view results in your client.

Typical workflows you can build: create a new expense with the add_expense action, fetch a specific expense with get_expense, view a filtered list with list_expenses, update an existing record with update_expense, remove entries with delete_expense or delete_expenses_by_date_range, summarize expenses by category with summarize, review overall statistics with get_statistics, and export all data with export_expenses.

How to install

Prerequisites you need before you begin:

Install Python on your system if it is not already present.
Install the uv tool used to run MCP servers and proxies.
Have Node.js available if you plan to work with claude-related tooling.

Follow these concrete steps to run Trackor locally:

uv sync
uv run main.py
uv run proxy/main.py

Additional setup notes

Inspector tool for testing JSON RPC calls can be used during development with the following command. It helps verify requests and responses against the MCP bridge.

uv run fastmcp dev .\main.py

Developer notes for local development and CLI tooling

If you are using Claude Desktop and want to load the local MCP server with a desktop extension, install the MCP packaging tool and create a trackor proxy package. This enables Claude to load the MCP server via the local STDIO proxy.

npm install -g @anthropic-ai/mcpb

mcpb pack proxy/ trackor-proxy.mcpb

# Then install the extension in Claude Desktop as an Advanced extension and load trackor-proxy.mcpb

Available tools

add_expense

Create a new expense entry with details like amount, category, date, and notes.

get_expense

Fetch a single expense by its ID to view details.

list_expenses

List expenses with optional filters such as date range, category, or amount.

update_expense

Modify fields of an existing expense entry.

delete_expense

Remove a specific expense by its ID.

delete_expenses_by_date_range

Delete all expenses falling within a specified date range.

summarize

Summarize expenses by category and subcategory to reveal spending patterns.

get_statistics

Return overall statistics and a monthly breakdown of expenses.

export_expenses

Export all expenses in JSON or CSV format for external usage.