The economic control and trust infrastructure for autonomous agents — enabling them to pay, get paid, and transact securely under real-time policy enforcement.
Every AI agent that touches money needs the same handful of things: wallet orchestration, payment routing, spending guardrails, trust evaluation, audit trails, and recovery flows. Today teams wire these by hand, one integration at a time, and re-learn the same compliance lessons the hard way.
OmniClaw replaces that patchwork with one SDK. You get guarded execution, policy enforcement, and regulatory-aware defaults out of the box — so you can focus on what your agent actually does, not how it moves money.
For Agents That Pay — guarded pay() execution, simulate() before funds move, x402 and direct transfer routing, cross-chain USDC flows, nanopayments via Circle Gateway
For Agents That Earn — Seller SDK, sell() decorator, facilitated transfers, trust-gated access
For Operators That Control — policy enforcement, spending limits, velocity controls, circuit breakers, audit-ready logs, recovery flows
pip install omniclawFor local development:
uv sync --extra devgit clone https://github.com/omnuron/omniclaw.git
cd omniclawCreate a .env file with your Circle credentials:
CIRCLE_API_KEY=your_circle_api_key
ENTITY_SECRET=your_entity_secret
Start the server:
docker-compose up -d
# Server runs at http://localhost:8088Verify the setup:
omniclaw doctor # Verify credentials
omniclaw env # List all env varspip install omniclaw
omniclaw-cli configure --server-url http://localhost:8088 --token <TOKEN> --wallet primaryfrom omniclaw import OmniClaw, Network
client = OmniClaw(network=Network.BASE_SEPOLIA)For detailed guides and architecture docs, see the Wiki:
| Page | Description |
|---|---|
| Getting Started | Full installation, environment setup, and first payment walkthrough |
| Architecture | System design, module breakdown, and payment flow |
| Compliance Design | Authorization traceability, regulatory alignment (CLARITY Act, GENIUS Act), and gray-zone analysis |
| Trust & ERC-8004 | Trust evaluation framework, on-chain signals, and audit comparison |
| API Reference | pay(), simulate(), sell(), NanoPayment, Trust, CLI |
| Contributing Guide | Dev setup, branch workflow, commit conventions, code quality |
See CONTRIBUTING.md and the Contributing Guide for development setup, coding standards, and how to submit pull requests.
MIT — see LICENSE for details.