task-manager-mcp
An MCP server that automates project task breakdown, dependency management, and smart task recommendations, integrating with LLMs like Gemini and OpenAI.
README
Task Manager MCP Service
An intelligent task management service based on Model Context Protocol (MCP), helping to automate project task breakdown, dependency management, and execution recommendations.
Features
- Automated Task Breakdown: Automatically extract and plan task structures from PRD documents
- Dependency Management: Intelligently handle dependencies between tasks, avoiding circular dependencies
- Smart Task Recommendations: Recommend the next task to execute based on dependency status and priority
- Subtask Expansion: Use LLM to automatically expand main tasks into detailed subtasks
- Code Association: Record associations between tasks and implementation code for better traceability
- Task Priority: Support multi-level task priorities and tag management
- MCP Integration: Native support for Model Context Protocol (MCP) for easy collaboration with LLMs
Quick Start
Installation
Recommended installation using uv:
# Install uv
pip install uv
# Install dependencies
uv pip install -r requirements.txt
Or use traditional pip installation:
pip install fastmcp uvicorn pydantic google-generativeai
Environment Configuration
Set necessary environment variables:
# Gemini configuration
export GEMINI_API_KEY="your-api-key-here"
export LLM_PROVIDER="gemini"
export MODEL_NAME="gemini-1.5-flash"
# Or OpenAI configuration
# export OPENAI_API_KEY="your-api-key-here"
# export LLM_PROVIDER="openai"
# export MODEL_NAME="gpt-4o"
# Optional: proxy settings
export HTTP_PROXY="http://your-proxy:port"
export HTTPS_PROXY="http://your-proxy:port"
# Optional: output directory
export MCP_OUTPUT_DIR="/path/to/output"
Basic Usage
- Start the service (using uv):
uv run --with fastmcp fastmcp run src/server.py
Or directly using Python:
cd src
python server.py
- Configure MCP service in Cursor IDE:
Edit the ~/.cursor/mcp.json file (usually located at C:\Users\<username>\.cursor\mcp.json), find the mcpServers section and add or update the task-manager configuration:
{
"mcpServers": {
"task-manager": {
"command": "uv",
"args": [
"run",
"--with",
"fastmcp",
"fastmcp",
"run",
"D:\\code\\git_project\\task-manager-mcp\\src\\server.py" // Note: Replace this with the absolute path to your local server.py
],
"env": {
"GEMINI_API_KEY": "<Your Gemini API Key>",
"HTTP_PROXY": "http://127.0.0.1:7890", // If proxy is needed
"HTTPS_PROXY": "http://127.0.0.1:7890", // If proxy is needed
"MODEL_NAME": "gemini-1.5-flash", // Or other supported models
"LLM_PROVIDER": "gemini", // Or openai
"MCP_OUTPUT_DIR": "D:\\path\\to\\your\\output\\directory\\" // Optional: specify output directory
}
}
// There might be other server configurations...
}
}
Note:
- Replace
D:\code\git_project\task-manager-mcp\src\server.pywith the absolute path to your localserver.pyfile. - Ensure the environment variables in
envare set correctly, especially the API key and proxy settings (if needed). MCP_OUTPUT_DIRis optional, used to specify the output location for task-related files (such as JSON, Markdown).
- Using the service in Cursor:
@task-manager decompose_prd prd_content="file:///D:/path/to/prd.md"
Documentation
For detailed documentation, please refer to the docs/ directory:
- Design Document - System design overview
- Getting Started - Getting started guide
- API Reference - Detailed API reference
- MCP Rules - LLM calling conventions
- Configuration Examples - Configuration examples
- Implementation Guide - Implementation guide
- Installation Guide - Detailed installation instructions
- To-Do List - Development roadmap
Key Functions
PRD Parsing and Task Breakdown
Automatically extract tasks and dependencies from project requirement documents:
@task-manager decompose_prd prd_content="file:///D:/path/to/prd.md"
Task Management
Create and update tasks (including status, dependencies, code references, etc.):
@task-manager add_task name="Implement login function" description="Implement user login function, including form validation" priority="high" tags="frontend,user function"
@task-manager update_task task_id="1" status="in_progress" dependencies="2,3"
@task-manager get_task task_id="1"
@task-manager get_task_list status="todo" priority="high" tag="frontend"
Subtask Expansion
Expand a main task into multiple subtasks:
@task-manager expand_task task_id="1" num_subtasks="5"
Smart Task Recommendation
Get the next task that should be executed:
@task-manager get_next_executable_task
Code Reference Management
Update code files associated with a task:
@task-manager update_task_code_references task_id="1" code_files="src/login.js,src/utils/validation.js"
System Architecture
The system is built on Python and the FastMCP framework, providing interfaces using MCP tools. The architecture includes:
- MCP Service Layer: Handles MCP protocol communication and tool invocation
- Core Logic Layer: Implements core features such as task management and dependency checking
- LLM Integration Layer: Integrates with LLM services like Gemini/OpenAI for intelligent parsing
- Storage Layer: Supports in-memory storage (default) and database storage (extensible)
Contributing
Contributions are welcome, including code contributions, issue reports, or new feature suggestions. Please refer to the Contributing Guide for details.
License
MIT License
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.