Backlog

Provides a minimal task backlog MCP server with create, read, update, and list capabilities.

typescript

0

GitHub Stars

typescript

Language

4 months ago

First Indexed

3 weeks 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": {
    "gkoreli-backlog-mcp": {
      "command": "node",
      "args": [
        "/path/to/backlog-mcp/dist/server.js"
      ]
    }
  }
}

This MCP server provides a minimal backlog for tracking tasks. It records task state and supports common backlog actions without enforcing a workflow, making it easy to integrate into your existing toolchain or automation flows.

How to use

You interact with the backlog MCP server through your MCP client to list tasks, view a task by ID, create new tasks, and update existing ones. You can filter the list by status (open, in_progress, blocked, done, cancelled) and request summaries to get quick counts. Tasks include fields like id, title, description, status, timestamps, and optional evidence. Use the update operation to change any mutable field such as title, description, status, blocked_reason, or evidence.

How to install

Prerequisites: ensure Node.js and npm are installed on your system.

Clone or download the backlog-mcp project to your environment and install dependencies.

Install dependencies, build, and start the server with these commands:

npm install
npm run build
npm start

Other setup and configuration

If you want to run the server locally from a specific path, you can configure your MCP client to launch the local runtime directly using the provided path to the compiled server. For example, you can point your client configuration to the server entry at the path shown below.

{
  "mcpServers": {
    "backlog": {
      "command": "node",
      "args": ["/path/to/backlog-mcp/dist/server.js"]
    }
  }
}

Notes on storage and license

The backlog data is stored in a single file at data/backlog.json. It uses atomic writes via a temporary file and rename to ensure data integrity.

License: MIT.

Available tools

backlog_list

List tasks with optional status filtering and summary mode to return counts.

backlog_get

Retrieve a single task by its ID.

backlog_create

Create a new task with a title, optional description, and initial status.

backlog_update

Update any field of an existing task, including title, description, status, blocked_reason, and evidence.