🚀 task-cli – AI-Driven CLI Project Management & Gantt Chart Tool

An open-source AI Agent that automatically decomposes tasks, generates Gantt charts, and provides project Q&A within your terminal.

English | 中文文档

In a nutshell
task-cli leverages an LLM to handle AI project management from the command line: automatically generating task lists, interactive Gantt charts, and real-time Q&A.

• 🤖 AI Task Decomposition – Based on high-level objectives, utilizes an LLM to automatically break down complex tasks into executable sub-tasks. Supports multi-turn dialogue to continuously refine the generated task plan.
• 📊 Interactive Gantt Charts – Automatically generates HTML Gantt charts, making task dependencies clear and intuitively tracking project progress.
• 💬 Intelligent Project Q&A – Chat with your project database using natural language queries.
• 🔧 Rich CLI Commands – init, decompose, gantt, report cover the entire project lifecycle.
• 🌍 Language Support – The agent can be configured to respond in the user's preferred language.

pip install task-cli-tool

# test
task-cli

Use the init command to start a new project:

task-cli init
? Enter project name: Cross-platform To-Do App
? Enter project goal: All software features are fully functional

Before use, you need to configure the DEEPSEEK_API_KEY environment variable to use the LLM's API.

After initialization, run the program to enter the assistant terminal:

set DEEPSEEK_API_KEY="your-api-key"
task-cli

Type / to trigger subcommand autocompletion.

Debug Mode: Set the environment variable DEBUG_MODE=true to view LLM request details.

> /decompose Develop a cross-platform to-do app
✅ Decomposed into 8 sub-tasks. View with /list

> /gantt
📊 Gantt chart saved to gantt.html

Open gantt.html to see the interactive timeline:

# Ask the assistant for guidance on executing a single task
/howto 

# Update task status
/update

# Project summary report
/report

# Ask any questions about the project in natural language
/qa

# Display help information
/help

All forms of contribution are welcome!

• Found a Bug → Submit an Issue
• Feature Suggestion → Start a Discussion
• Code Contribution → Submit a Pull Request

How to dev：

# Clone the repository
git clone https://github.com/sunjiawe/task-cli.git
cd task-cli

# Create a virtual environment (optional)
python -m venv venv && source venv/bin/activate  # Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt
pip install -e .
task-cli

This project is licensed under the Apache 2.0 License. Apache 2.0 © 2025 sunjiawe

• Pocket Flow - The minimalist Agent framework that drives this project.
• gemini-cli - An open-source AI agent that brings the power of Gemini directly into your terminal.