LICENSE Documentation

Models and Providers Copy Section

• Role-based model setup: main, research, fallback
• Provider selection (Anthropic, OpenAI, Perplexity, etc.)
• Model IDs per role
• Temperature, max tokens, and other generation settings
• Custom base URLs for OpenAI-compatible APIs

Global Settings Copy Section

• logLevel: Logging verbosity
• debug: Enable/disable debug mode
• projectName: Optional name for your project
• defaultTag: Default tag for task grouping
• defaultSubtasks: Number of subtasks to auto-generate
• defaultPriority: Priority level for new tasks

API Endpoint Overrides Copy Section

• ollamaBaseURL: Custom Ollama server URL
• azureBaseURL: Global Azure endpoint
• vertexProjectId: Google Vertex AI project ID
• vertexLocation: Region for Vertex AI models

Tag and Git Integration Copy Section

• Default tag context per project
• Support for task isolation by tag
• Manual tag creation from Git branches

State Management Copy Section

• Active tag tracking
• Migration state
• Last tag switch timestamp

MCP (Model Control Protocol) lets you run Task Master directly from your editor.

1. Add your MCP config at the following path depending on your editor Copy Chapter

Manual Configuration Copy Chapter

Cursor & Windsurf (`mcpServers`) Copy Section

{
  "mcpServers": {
    "taskmaster-ai": {
      "command": "npx",
      "args": ["-y", "--package=task-master-ai", "task-master-ai"],
      "env": {
        "ANTHROPIC_API_KEY": "YOUR_ANTHROPIC_API_KEY_HERE",
        "PERPLEXITY_API_KEY": "YOUR_PERPLEXITY_API_KEY_HERE",
        "OPENAI_API_KEY": "YOUR_OPENAI_KEY_HERE",
        "GOOGLE_API_KEY": "YOUR_GOOGLE_KEY_HERE",
        "MISTRAL_API_KEY": "YOUR_MISTRAL_KEY_HERE",
        "OPENROUTER_API_KEY": "YOUR_OPENROUTER_KEY_HERE",
        "XAI_API_KEY": "YOUR_XAI_KEY_HERE",
        "AZURE_OPENAI_API_KEY": "YOUR_AZURE_KEY_HERE",
        "OLLAMA_API_KEY": "YOUR_OLLAMA_API_KEY_HERE"
      }
    }
  }
}

🔑 Replace YOUR_…_KEY_HERE with your real API keys. You can remove keys you don't use.

Note: If you see 0 tools enabled in the MCP settings, try removing the --package=task-master-ai flag from args.

VS Code (`servers` + `type`) Copy Section

{
  "servers": {
    "taskmaster-ai": {
      "command": "npx",
      "args": ["-y", "--package=task-master-ai", "task-master-ai"],
      "env": {
        "ANTHROPIC_API_KEY": "YOUR_ANTHROPIC_API_KEY_HERE",
        "PERPLEXITY_API_KEY": "YOUR_PERPLEXITY_API_KEY_HERE",
        "OPENAI_API_KEY": "YOUR_OPENAI_KEY_HERE",
        "GOOGLE_API_KEY": "YOUR_GOOGLE_KEY_HERE",
        "MISTRAL_API_KEY": "YOUR_MISTRAL_KEY_HERE",
        "OPENROUTER_API_KEY": "YOUR_OPENROUTER_KEY_HERE",
        "XAI_API_KEY": "YOUR_XAI_KEY_HERE",
        "AZURE_OPENAI_API_KEY": "YOUR_AZURE_KEY_HERE"
      },
      "type": "stdio"
    }
  }
}

🔑 Replace YOUR_…_KEY_HERE with your real API keys. You can remove keys you don't use.

2. (Cursor-only) Enable Taskmaster MCP

Open Cursor Settings (Ctrl+Shift+J) ➡ Click on MCP tab on the left ➡ Enable task-master-ai with the toggle

3. (Optional) Configure the models you want to use

In your editor's AI chat pane, say:

Change the main, research and fallback models to <model_name>, <model_name> and <model_name> respectively.

For example, to use Claude Code (no API key required):

Change the main model to claude-code/sonnet

4. Initialize Task Master

In your editor's AI chat pane, say:

Initialize taskmaster-ai in my project

CLI Installation Copy Chapter

# Install globally
npm install task-master-ai

# OR install locally within your project
npm install task-master-ai

Initialize a new project Copy Chapter

# If installed globally
task-master init

# If installed locally
npx task-master init

# Initialize project with specific rules
task-master init --rules cursor,windsurf,vscode

This will prompt you for project details and set up a new project with the necessary files and structure.

File: docs/getting-started/quick-start/prd-quick.mdx Copy Chapter

title: PRD Creation and Parsing sidebarTitle: "PRD Creation and Parsing" Copy Chapter

Writing a PRD

A PRD (Product Requirements Document) is the starting point of every task flow in Task Master. It defines what you're building and why. A clear PRD dramatically improves the quality of your tasks, your model outputs, and your final product — so it’s worth taking the time to get it right.

What Makes a Good PRD? Copy Chapter

• Clear objective — what’s the outcome or feature?
• Context — what’s already in place or assumed?
• Constraints — what limits or requirements need to be respected?
• Reasoning — why are you building it this way?

The more context you give the model, the better the breakdown and results.

Writing a PRD for Task Master Copy Chapter

An example PRD can be found in .taskmaster/templates/example_prd.txt

You can co-write your PRD with an LLM model using the following workflow:

1. Chat about requirements — explain what you want to build.
2. Show an example PRD — share the example PRD so the model understands the expected format. The example uses formatting that work well with Task Master's code. Following the example will yield better results.
3. Iterate and refine — work with the model to shape the draft into a clear and well-structured PRD.

This approach works great in Cursor, or anywhere you use a chat-based LLM.

Where to Save Your PRD Copy Chapter

Place your PRD file in the .taskmaster/docs folder in your project.

• You can have multiple PRDs per project.
• Name your PRDs clearly so they’re easy to reference later.
  • Examples: dashboard_redesign.txt, user_onboarding.txt

Parse your PRD into Tasks

This is where the Task Master magic begins.

In Cursor's AI chat, instruct the agent to generate tasks from your PRD:

Please use the task-master parse-prd command to generate tasks from my PRD. The PRD is located at .taskmaster/docs/.txt.

The agent will execute the following command which you can alternatively paste into the CLI:

task-master parse-prd .taskmaster/docs/.txt

This will:

• Parse your PRD document
• Generate a structured tasks.json file with tasks, dependencies, priorities, and test strategies

Now that you have written and parsed a PRD, you are ready to start setting up your tasks.

File: docs/getting-started/quick-start/quick-start.mdx Copy Chapter

title: Quick Start sidebarTitle: "Quick Start" Copy Chapter

This guide is for new users who want to start using Task Master with minimal setup time.

It covers:

• Requirements: You will need Node.js and an AI model API Key.
• Installation: How to Install Task Master.
• Configuration: Setting up your API Key, MCP, and more.
• PRD: Writing and parsing your first PRD.
• Task Setup: Preparing your tasks for execution.
• Executing Tasks: Using Task Master to execute tasks.
• Rules & Context: Learn how and why to build context in your project over time.

File: docs/getting-started/quick-start/requirements.mdx Copy Chapter

title: Requirements sidebarTitle: "Requirements" Copy Chapter

Before you can start using TaskMaster AI, you'll need to install Node.js and set up at least one model API Key.

1. Node.js Copy Chapter

TaskMaster AI is built with Node.js and requires it to run. npm (Node Package Manager) comes bundled with Node.js.

choco install nodejs

winget install OpenJS.NodeJS

2. Model API Key Copy Chapter

Taskmaster utilizes AI across several commands, and those require a separate API key. For the purpose of a Quick Start we recommend setting up an API Key with Anthropic for your main model and Perplexity for your research model (optional but recommended).

Task Master shows API costs per command used. Most users load $5-10 on their keys and don't have to top it off for a few months.

At least one (1) of the following is required:

1. Anthropic API key (Claude API) - recommended for Quick Start
2. OpenAI API key
3. Google Gemini API key
4. Perplexity API key (for research model)
5. xAI API Key (for research or main model)
6. OpenRouter API Key (for research or main model)
7. Claude Code (no API key required - requires Claude Code CLI)

File: docs/getting-started/quick-start/tasks-quick.mdx Copy Chapter

title: Tasks Setup sidebarTitle: "Tasks Setup" Copy Chapter

Now that your tasks are generated you can review the plan and prepare for execution.

Expand Tasks Copy Chapter

Used to add detail to tasks and create subtasks. We recommend expanding all tasks using the MCP request below:

Expand all tasks into subtasks.

The agent will execute

task-master expand --all

List/Show Tasks Copy Chapter

Used to view task details. It is important to review the plan and ensure it makes sense in your project. Check for correct folder structures, dependencies, out of scope subtasks, etc.

To see a list of tasks and descriptions use the following command:

List all pending tasks so I can review.

To see all tasks in the CLI you can use:

task-master list

To see all implementation details of an individual task, including subtasks and testing strategy, you can use Show Task:

Show task 2 so I can review.

task-master show --id=<##>

Update Tasks Copy Chapter

If the task details need to be edited you can update the task using this request:

Update Task 2 to use Postgres instead of MongoDB and remove the sharding subtask

Or this CLI command:

task-master update-task --id=2 --prompt="use Postgres instead of MongoDB and remove the sharding subtask"

Analyze complexity Copy Chapter

Task Master can provide a complexity report which can be helpful to read before you begin. If you didn't already expand all your tasks, it could help identify which could be broken down further with subtasks.

Can you analyze the complexity of our tasks to help me understand which ones need to be broken down further?

You can view the report in a friendly table using:

Can you show me the complexity report in a more readable format?

Now you are ready to begin executing tasks