Neural SDK

Deterministic prediction-market kernel for portable contracts, normalization, replay, and paper-first integration beneath Vaticor.

Documentation | Examples | Contributing

Stable Kernel

The dependency-free stable surface lives at neural.kernel. Provider access, strategy analysis, deployment, sentiment, and FIX helpers remain experimental or deprecated compatibility modules.

Install

Using uv:

uv add neural-sdk
uv add "neural-sdk[trading]"

Using pip:

pip install neural-sdk
pip install "neural-sdk[trading]"

Verify the stable installed surface without credentials or network access:

neural --json capabilities
neural --json replay demo

See Stable Kernel and Compatibility for the capability matrix, extras, and deprecation policy.

CLI Bridge

The base install ships a neural CLI intended to be the stable machine-readable bridge for the TypeScript Neural TUI.

neural doctor
neural --json capabilities
neural --json providers list

Current bridge commands:

• doctor
• capabilities
• replay demo
• providers list
• markets list
• quote
• positions
• paper order
• deployments list
• deployments status
• deployments logs
• deployments stop

Highlighted Examples

• examples/09_daily_nba_markets.py: public Kalshi NBA market snapshot with sports-aware parsing.
• examples/10_daily_nba_markets_polymarket.py: public Polymarket US NBA market snapshot using the normalized market interface.
• examples/12_cross_provider_analysis_bridge.py: live cross-provider normalization plus one shared analysis pass.
• examples/13_deterministic_cross_provider_replay.py: reproducible replay proving the analysis stack generates buy_yes, buy_no, and hold decisions on standardized cross-provider data.

Credentials

Create a .env file with your Kalshi credentials:

KALSHI_API_KEY_ID=your_api_key_id
KALSHI_PRIVATE_KEY_BASE64=base64_encoded_private_key
KALSHI_ENV=prod

The SDK loads credentials from environment variables or your local secrets files.

Development

git clone https://github.com/IntelIP/Neural.git
cd neural
uv sync --extra dev --extra trading --extra sentiment --extra analysis --extra deployment
uv run pytest
uv run ruff check .
uv run black --check .

Testing

uv run pytest
uv run pytest --cov=neural tests/

Resources

• Documentation: docs/
• Examples: examples/
• Issues: GitHub Issues
• Discussions: GitHub Discussions

License

This project is licensed under the MIT License. See LICENSE.