M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
C

Claude Task Master

Community
An AI-powered task management system that drops into Cursor, Lovable, Windsurf, Roo, and more.
How to install ↓ Compare GitHub source repository ↗
Category
Dev Tools #290 of 438
Stars
★ 28.2k Very popular
Transport
stdio (local process)
Runtime
Node.js
Credentials
API key / credential required
License
Other / unspecified
Last commit
Tools
15
47FMRS · D

Task Master is a feature-rich AI task management MCP server supporting multiple AI providers and editors, offering a complete workflow from PRD to task tracking. Its strengths are easy integration and diverse tools, but its non-official nature, license restrictions, and reliance on external AI services are factors to consider.

Strongest · Maintenance 15/20 Weakest · Reliability 7/20

Reliability
7/20
Security and permissions
8/20
Maintenance
15/20
Documentation
9/20
Setup experience
8/20
Why each score
Reliability 7/20
Evidence: README claims 36 MCP tools with configurable subsets via TASK_MASTER_TOOLS, and CI exists. Deductions: static review did not execute the server; no committed server-level integration tests prove the MCP init/tool-list handshake; tool/behavior parity between README and source is unverified; error-handling paths are not confirmed. Thus score is capped below 12 and set conservatively.
Security and permissions 8/20
Evidence: API keys are passed via environment variables; install examples use placeholders only, with no real tokens; no obvious exfiltration or malware patterns found statically. Deductions: least-privilege scope is not fully articulated; destructive/state-changing operations (e.g., task deletion, moves) lack explicit confirmation flows; external data flow and network calls are not fully disclosed. No red-line triggered, but not a full score.
Maintenance 15/20
Evidence: Repository is active with many stars and open issues, a CI workflow badge and workflow file exist, and the license block (MIT with Commons Clause) matches the README badge; issue response appears ongoing. Deductions: static review cannot verify release cadence, dependency update hygiene, or a security-response channel; license is non-standard, and NOASSERTION in metadata adds minor uncertainty.
Documentation 9/20
Evidence: README covers install (MCP and CLI), API-key/auth setup for multiple providers, model configuration, tool-loading modes, common commands, and a troubleshooting section. Deductions: limitations and costs are not disclosed; troubleshooting is thin; tool parameter details are only claimed, not verified against source; docs rely on the untrusted README rather than committed code-level documentation.
Setup experience 8/20
Evidence: README includes per-editor config examples (Cusor, Windsurf, VS Code, Q CLI) and a one-line Claude Code command, and the API-key prerequisites are described explicitly. Deductions: setup is manual and editor-specific with fragile env-var configuration; there is no committed integration test demonstrating a working connection; static review cannot confirm that npx -y task-master-ai starts cleanly, so setup is capped below 15.

Static review · not runListed 2026-08-07

Read the FMRS scoring method →

Fit and risk

What it can accessReads local filesWrites / deletes local filesRuns commands or codeUses the network

Best for

  • Product managers and developers who want to manage development tasks directly from AI chat
  • Users of AI editors like Cursor, Windsurf, VS Code
  • Teams needing to automate task generation from requirements documents (PRDs)
  • Developers looking to minimize context window usage and optimize AI tool loading

Not for

  • Traditional project management scenarios without AI assistance
  • Environments without any API key and without Claude Code or Codex CLI installed
  • Users seeking fully offline, dependency-free task management
  • Those who need commercial hosting or want to resell the software (license limitations)

Required permissions

  • Read project files (e.g., PRD documents)
  • Create and modify task files in the project directory (e.g., .taskmaster folder)
  • Execute the task-master-ai package via npx
  • Access external AI APIs (e.g., Anthropic, OpenAI) to generate and analyze tasks

Risks and side effects

  • AI-generated tasks may be inaccurate and require human review
  • Project file data is sent to external AI services, posing potential data leakage risks
  • API keys in config files could be leaked and misused
  • Non-official server with MIT with Commons Clause license; resale or hosting is prohibited
  • Large default toolset (36) may increase context consumption; configuration optimization recommended

Setup

Before you start

Runtime：Node.js

ANTHROPIC_API_KEY optionalsecret Anthropic Claude API key from console.anthropic.com, used for the main model (e.g. Claude 3.7).
OPENAI_API_KEY optionalsecret OpenAI API key from platform.openai.com, optional main/research provider.
GOOGLE_API_KEY optionalsecret Google Gemini API key from Google AI Studio, optional provider.
PERPLEXITY_API_KEY optionalsecret Perplexity API key, used for the research model; highly recommended.
OPENROUTER_API_KEY optionalsecret OpenRouter API key for multi-model access, optional.
XAI_API_KEY optionalsecret xAI (Grok) API key, usable for main or research model, optional.
MISTRAL_API_KEY optionalsecret Mistral API key, optional model provider.
GROQ_API_KEY optionalsecret Groq API key, optional model provider.
AZURE_OPENAI_API_KEY optionalsecret Azure OpenAI service key, optional model provider.
OLLAMA_API_KEY optionalsecret Ollama key for local models, optional provider.
Other optional settings (1)
TASK_MASTER_TOOLS optional Controls which MCP tools load to reduce context usage; options: all (default), standard, core/lean, or a comma-separated tool list.
  1. Add the install_config above to your MCP configuration file (e.g., ~/.cursor/mcp.json for Cursor, .vscode/mcp.json for VS Code, etc.).
  2. Replace placeholder API keys with your actual provider keys (at least one is required unless using Claude Code or Codex CLI).
  3. Restart your editor and enable the MCP server.
  4. In your AI chat, run 'Initialize taskmaster-ai in my project' to initialize.
claude_desktop_config.json 
{
  "mcpServers": {
    "task-master-ai": {
      "command": "npx",
      "args": [
        "-y",
        "task-master-ai"
      ],
      "env": {
        "ANTHROPIC_API_KEY": "YOUR_ANTHROPIC_API_KEY_HERE",
        "OPENAI_API_KEY": "YOUR_OPENAI_KEY_HERE",
        "GOOGLE_API_KEY": "YOUR_GOOGLE_KEY_HERE",
        "PERPLEXITY_API_KEY": "YOUR_PERPLEXITY_API_KEY_HERE",
        "OPENROUTER_API_KEY": "YOUR_OPENROUTER_KEY_HERE",
        "XAI_API_KEY": "YOUR_XAI_KEY_HERE",
        "MISTRAL_API_KEY": "YOUR_MISTRAL_KEY_HERE",
        "GROQ_API_KEY": "YOUR_GROQ_KEY_HERE",
        "AZURE_OPENAI_API_KEY": "YOUR_AZURE_KEY_HERE",
        "OLLAMA_API_KEY": "YOUR_OLLAMA_API_KEY_HERE"
      }
    }
  }
}

Shown for Claude Desktop. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.

.vscode/mcp.json
{
  "servers": {
    "task-master-ai": {
      "command": "npx",
      "args": [
        "-y",
        "task-master-ai"
      ],
      "env": {
        "ANTHROPIC_API_KEY": "YOUR_ANTHROPIC_API_KEY_HERE",
        "OPENAI_API_KEY": "YOUR_OPENAI_KEY_HERE",
        "GOOGLE_API_KEY": "YOUR_GOOGLE_KEY_HERE",
        "PERPLEXITY_API_KEY": "YOUR_PERPLEXITY_API_KEY_HERE",
        "OPENROUTER_API_KEY": "YOUR_OPENROUTER_KEY_HERE",
        "XAI_API_KEY": "YOUR_XAI_KEY_HERE",
        "MISTRAL_API_KEY": "YOUR_MISTRAL_KEY_HERE",
        "GROQ_API_KEY": "YOUR_GROQ_KEY_HERE",
        "AZURE_OPENAI_API_KEY": "YOUR_AZURE_KEY_HERE",
        "OLLAMA_API_KEY": "YOUR_OLLAMA_API_KEY_HERE"
      }
    }
  }
}

Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).

Terminal
claude mcp add task-master-ai -e ANTHROPIC_API_KEY=YOUR_ANTHROPIC_API_KEY_HERE -e OPENAI_API_KEY=YOUR_OPENAI_KEY_HERE -e GOOGLE_API_KEY=YOUR_GOOGLE_KEY_HERE -e PERPLEXITY_API_KEY=YOUR_PERPLEXITY_API_KEY_HERE -e OPENROUTER_API_KEY=YOUR_OPENROUTER_KEY_HERE -e XAI_API_KEY=YOUR_XAI_KEY_HERE -e MISTRAL_API_KEY=YOUR_MISTRAL_KEY_HERE -e GROQ_API_KEY=YOUR_GROQ_KEY_HERE -e AZURE_OPENAI_API_KEY=YOUR_AZURE_KEY_HERE -e OLLAMA_API_KEY=YOUR_OLLAMA_API_KEY_HERE -- npx -y task-master-ai

Run it in a terminal; replace any <…> placeholders with your own values first.

Generate and validate client config

Check that it works

Check your editor's MCP settings to confirm the task-master-ai server is enabled with tools listed (if it shows 0 tools enabled, restart the editor and verify API keys), then ask "Initialize taskmaster-ai in my project" in chat and watch for tool calls.

Troubleshooting

  1. If `task-master init` doesn't respond, try running `node node_modules/claude-task-master/scripts/init.js` directly
  2. If you see '0 tools enabled' in MCP settings, restart your editor and check API key configuration
  3. Ensure the TASK_MASTER_TOOLS environment variable is set correctly (all, standard, core, etc.)
  4. Check network connectivity to ensure access to AI APIs
  5. Verify write permissions in the project directory

Things to try

Once connected, you can ask your AI assistant things like:

  • Initialize taskmaster-ai in my project
  • Can you parse my PRD at scripts/prd.txt?
  • What's the next task I should work on?
  • Can you help me expand task 4?

Tools 15

get_tasks read-only
Get the current list of tasks.
next_task read-only
Get the next task to work on.
get_task read-only
Get details of a single task.
set_task_status writes
Set task status (e.g., pending, in-progress, done).
update_subtask writes
Update subtask status.
parse_prd writes
Parse a PRD and generate a task list.
expand_task writes
Expand a task into more detailed subtasks.
initialize_project writes
Initialize the Task Master project structure.
Show 7 more tools
analyze_project_complexity read-only
Analyze project complexity.
expand_all writes
Expand all tasks.
add_subtask writes
Add a subtask to a task.
remove_task destructive
Remove a task.
generate writes
Generate tasks.
add_task writes
Add a new task.
complexity_report read-only
Generate a complexity report.

Use cases

Automatically generate task lists from a PRD
Manage development tasks in your IDE via natural language
Track task status, dependencies, and tags
Use AI to research latest technical practices
Collaborate on AI-driven development workflows in teams

Supported clients

Claude Desktop
Cursor
Windsurf
VS Code
Q CLI

Listed from the project's documentation, not tested by this site.

Overview

Task Master (task-master-ai) is an AI-driven task management system designed for AI-assisted development, integrating seamlessly with editors like Cursor, Lovable, Windsurf, and Roo. It provides task management tools via MCP, supporting parsing requirements from PRDs, generating task lists, tracking progress, managing dependencies, and leveraging AI for research and loop execution. The system supports multiple AI providers (Anthropic, OpenAI, Google, Perplexity, etc.) and allows optimizing tool loading through environment variables (e.g., TASK_MASTER_TOOLS) to reduce context window usage.

Similar servers

Context7 80 · B

Upstash's official server providing up-to-date third-party library docs for AI coding assistants

★ 62.9k · Tools 2 Compare with this →
Compare with other MCP servers

Source revision c0c98d367c55 Data synced 2026-10-12 Read the FMRS scoring method