Project Tracking

Provides project and task management via MCP with a single SQLite database for persistent storage.

python

0

GitHub Stars

python

Language

4 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": {
    "leonj1-project-tracking-mcp": {
      "command": "python",
      "args": [
        "/path/to/project-tracking-mcp/server.py"
      ]
    }
  }
}

You run a lightweight MCP server that uses FastMCP to expose project and task management through a single SQLite file. This server lets you create projects, attach tasks with categories, track statuses, and store everything in a portable database for easy CRUD operations.

How to use

To interact with this MCP server, use an MCP client that communicates via the MCP protocol. You can start the server locally and connect a client to it, perform project CRUD operations, manage tasks within projects, and view aggregate statistics. The available actions include listing projects, creating new projects, retrieving full project details with tasks, deleting projects, creating tasks, updating task statuses, deleting tasks, and querying overall statistics.

How to install

Prerequisites check before you begin: Python 3.10 or newer, and a compatible MCP client environment. You will also need a way to run Python scripts (the system you’re using should have Python installed and accessible via the command line). Also ensure you have a working internet connection to install dependencies.

Step 1: Clone the project locally and move into its directory.

Step 2: Install dependencies using one of these methods.

Option A — Using pip (recommended for Python environments):

pip install -r requirements.txt

Option B — Using uvx (fast MCP workflow):

uv pip install fastmcp pydantic

Option C — Install directly without a file:

pip install fastmcp pydantic


Step 3: Start the MCP server.

python server.py

## Configuration and usage notes

The server stores all data in a single SQLite file named `projects.db`. It initializes automatically on first run, and you can perform full CRUD operations for both projects and tasks.

If you want to use the server with Claude Desktop or Claude Code, you can configure an MCP client to launch the server in MCP mode. The following configuration examples show how to point the MCP client to the local Python server and enable MCP-only mode.

## Additional configuration and examples

Configure Claude Desktop to connect to the MCP server by providing the following MCP entry. This runs the server script directly from its path.

{ "mcpServers": { "project-tracker": { "command": "python", "args": ["/path/to/project-tracking-mcp/server.py"] } } }


Alternate integration with Claude Code uses the MCP-only flag to run in MCP mode. The typical entry looks like this.

{ "project-tracker": { "command": "python", "args": ["/absolute/path/to/project-tracking-mcp/server.py", "--mcp-only"], "env": {} } }


## Available tools

### list\_projects

List all projects with summary information.

### create\_project

Create a new project with a name and optional description.

### get\_project

Get detailed information for a project, including all tasks.

### delete\_project

Delete a project and all of its tasks.

### create\_task

Add a task to a specific project with a description and category.

### update\_task\_status

Update the status of a task (backlog, in\_progress, review, complete).

### delete\_task

Delete a task.

### get\_project\_stats

Get overall statistics about projects and tasks.