A complete end-to-end AI agent similar to Manus AI for personal use.
This Personal AI Agent is a powerful AI assistant that can help you with a wide range of tasks, including:
- Information gathering, fact-checking, and documentation
- Data processing, analysis, and visualization
- Writing multi-chapter articles and in-depth research reports
- Creating websites, applications, and tools
- Using programming to solve various problems
- Automating processes
- Various tasks that can be accomplished using computers and the internet
The agent is powered by DeepSeek-R1, a state-of-the-art language model, and includes a comprehensive set of tools for interacting with your computer and the internet.
- Modular Architecture: Clearly defined components with clean interfaces
- Advanced Planning: Sophisticated planning capabilities using Chain of Thought, Tree of Thoughts, and ReAct
- Knowledge Management: Short-term and long-term memory with vector database integration
- Comprehensive Tool Integration: Shell, file, browser, information, and message tools
- Multiple Deployment Options: Docker, Google Colab Pro with TPU optimization, and local installation
- Unlimited Context: Knowledge module handles unlimited context through vector database integration
- TPU Optimization: Enhanced performance with TPU acceleration in Google Colab Pro
- User Guide: Complete guide to using your Personal AI Agent
- Technical Documentation: In-depth technical details about the system
- Installation Guide: Step-by-step installation instructions
- Make sure Docker and Docker Compose are installed
- Run the setup script:
./setup.sh- Edit the
.envfile to add your DeepSeek API key - Start the containers:
docker-compose up -d- Access the agent at http://localhost:8501
- Upload the
colab_deploy.shscript to your Google Colab notebook - Run the script:
!chmod +x colab_deploy.sh
!./colab_deploy.sh- Follow the prompts to configure your agent
- Start the agent:
%run /content/personal-ai-agent/launch.py --api-key YOUR_DEEPSEEK_API_KEY- Install the required packages:
pip install -r requirements.txt- Configure your environment:
cp config.yaml.example config/config.yaml
# Edit config/config.yaml to add your API keys- Start the agent:
./run.sh- Docker Engine (version 20.10.0 or higher)
- Docker Compose (version 2.0.0 or higher)
- 4GB RAM minimum (8GB recommended)
- 10GB free disk space
- Google Colab Pro subscription
- Google Drive account
- Python 3.10 or higher
- pip package manager
- Git
- 8GB RAM minimum
- 10GB free disk space
The Personal AI Agent is built on a modular architecture with several key components:
- Agent Core: Central coordinator that manages the agent loop, event processing, and state management
- Planning Module: Breaks down complex tasks into steps and monitors execution
- Knowledge Module: Manages memory and knowledge retrieval
- LLM Integration: Provides a unified interface to language models
- Tool Framework: Enables the agent to interact with external systems
For more details, see the Technical Documentation.
To verify that your installation is working correctly:
./test.shFor integration tests:
./integration_test.shYou can customize your Personal AI Agent by editing the configuration file:
- Docker:
config/config.yaml - Google Colab:
/content/personal-ai-agent/config/colab_config.yaml - Local:
config/config.yaml
See the User Guide for more details on customization options.
This project is for personal use only. It is not affiliated with Manus AI.
This project was inspired by Manus AI and leverages several open-source technologies:
- DeepSeek-R1 language model
- Qdrant vector database
- FastAPI and Uvicorn for API services
- Various Python libraries for tool integration
For issues and questions, please refer to the documentation:
If you encounter issues not covered in the documentation, check the logs:
- Docker:
docker-compose logs agent - Local: Check
logs/agent.log - Colab: Check the output in your notebook