Taskmaster

Taskmaster MCP Server is a lightweight, HTTP-enabled processor that lets AI agents manage tasks through a simple, four-step workflow. It keeps session and task state persistent, auto-assigns essential tools, and works with any language model to orchestrate straightforward task execution without unnecessary complexity.

How to use

Set up a session for a project, define the tasks you want to accomplish, fetch the next task to work on, and mark tasks as complete until you finish the session. You interact with the server over HTTP for remote control or run it locally via Python for development and testing. Your client will receive structured guidance on what to do next and how to proceed when a task is completed.

How to install

# Prerequisites
- Python 3.8+
- Git

# Clone the project and install dependencies
git clone https://example.com/taskmaster.git
cd taskmaster
pip install -r requirements.txt

# Run the server locally
python server.py

Configuration and usage notes

By default the server runs on port 8080 and exposes the MCP endpoint at /mcp. You can adjust the port with the PORT environment variable when needed. The system supports two primary connection methods: an HTTP remote server URL for clients and a local stdio command for running the server directly. The environment supports Smithery deployment via SMITHERY_DEPLOY for production workflows.

Additional notes

Key configuration snippets, deployment options, and troubleshooting guidance are provided to help you operate reliably in development and production environments. You will define sessions, create task lists, execute and complete tasks, and end sessions when your work is finished. The design emphasizes simplicity, linear progression, and persistent state so you always know what comes next.

Available tools

read_file

Read contents from a file in the workspace and return the data to the MCP client.

edit_file

Modify the contents of a file with provided updates, returning success status and any diffs.

run_terminal_cmd

Execute a terminal command on the host environment and return stdout/stderr results.

codebase_search

Search the codebase for matching symbols, definitions, or text, returning relevant results to guide task execution.