A specialized Bittensor subnet for AI agents across 14 different categories, providing decentralized AI services with category-specific optimization.

• Python 3.8+
• Bittensor wallet with TAO balance
• Git

git clone https://github.com/Agents-Labs/agent4all-bittensor-subnet.git
cd agent4all-bittensor-subnet
pip install -r requirements.txt

# Launch a data analyst miner
python scripts/launch_miner.py --category data-analyst

# Launch a finance miner  
python scripts/launch_miner.py --category finance

# Launch any other category
python scripts/launch_miner.py --category <category-name>

# Launch a data analyst validator
python scripts/launch_validator.py --category data-analyst

# Launch a finance validator
python scripts/launch_validator.py --category finance

# Launch any other category
python scripts/launch_validator.py --category <category-name>

agent4all-bittensor-subnet/
├── neurons/                    # Core Bittensor neurons
│   ├── miner.py               # Base miner implementation
│   ├── validator.py           # Base validator implementation
│   └── dataset.py             # Dataset handling
├── template/                   # Bittensor protocol templates
│   ├── protocol.py            # Protocol definitions
│   ├── base/                  # Base classes
│   ├── validator/             # Validator templates
│   └── utils/                 # Utility functions
├── scripts/                    # Category-specific scripts
│   ├── category_miners/       # 14 miner scripts (one per category)
│   ├── category_validators/   # 14 validator scripts (one per category)
│   ├── utils/                 # Shared utilities
│   ├── launch_miner.py        # Miner launcher
│   ├── launch_validator.py    # Validator launcher
│   └── README.md              # Scripts documentation
├── category_registry.py        # Category definitions
├── config.yaml                # Configuration file
├── pyproject.toml             # Python dependencies
├── requirements.txt           # Requirements
├── README.md                  # This file
├── LICENSE                    # License
└── .gitignore                # Git ignore rules

The subnet supports 14 specialized AI agent categories:

1. data-analyst - Data analysis and visualization
2. finance - Financial analysis and investment advice
3. image - Image generation and processing
4. image-to-text - Image captioning and OCR
5. markdown - Markdown processing and documentation
6. planning - Project planning and scheduling
7. programming - Code generation and review
8. recommendation - Product and content recommendation
9. research - Research and information gathering
10. shopping - E-commerce and shopping assistance
11. single - General purpose conversation
12. video - Video processing and analysis
13. websearch - Web search and browsing
14. wikipedia - Wikipedia content processing

Edit config.yaml to configure:

• Network settings (testnet/mainnet)
• Wallet configuration
• Neuron parameters
• Category weights

# Create wallet
btcli wallet new_coldkey --wallet.name agent4all_coldkey
btcli wallet new_hotkey --wallet.name agent4all_coldkey --wallet.hotkey <category>_hotkey

# Register on subnet
btcli subnet register --wallet.name agent4all_coldkey --wallet.hotkey <category>_hotkey --subtensor.network test

Each category has specialized:

• Task processing optimized for the domain
• Evaluation metrics specific to the category
• Performance tracking for monitoring
• Scoring algorithms for fair rewards

1. Update category_registry.py
2. Create miner script in scripts/category_miners/
3. Create validator script in scripts/category_validators/
4. Update configuration

• Modify task handlers in miners
• Adjust evaluation criteria in validators
• Configure scoring weights
• Add category-specific metrics

• Response times
• Success rates
• Score distributions
• Error tracking

All scripts support configurable logging:

--logging.level DEBUG|INFO|WARNING|ERROR

• Testnet: --subtensor.network test (default)
• Mainnet: --subtensor.network main
• Local: --subtensor.network local

1. Wallet errors: Check balance and registration
2. Network errors: Verify connectivity
3. Import errors: Install dependencies
4. Permission errors: Check file permissions

• Check logs for error messages
• Verify configuration parameters
• Test with different networks
• Consult script documentation

• Scripts Documentation - Detailed script usage
• Technical Whitepaper - Technical details
• Setup Guide - Complete setup instructions
• Project Summary - Cleanup and structure overview

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Test thoroughly
5. Submit a pull request

# Clone the repository
git clone https://github.com/Agents-Labs/agent4all-bittensor-subnet.git

# Install dependencies
pip install -r requirements.txt

# Run tests (if available)
python -m pytest tests/

# Make your changes and submit a PR

This project is licensed under the MIT License - see the LICENSE file for details.

• Bittensor Foundation for the protocol
• Agents-Labs for the project
• Community contributors
• AI research community

Agent4All Bittensor Subnet - Decentralized AI for everyone, specialized for every domain.

Repository: https://github.com/Agents-Labs/agent4all-bittensor-subnet