We are thrilled to introduce GoHyperrr, a next-generation commerce framework engineered specifically for the era of agentic workflows and machine-to-machine transactions.
As AI agents increasingly act as decision-makers and purchasers, traditional storefronts and checkout structures are no longer sufficient. GoHyperrr bridges this gap by decoupling the core commerce domain logic from the execution engine, exposing transactional nodes directly as Model Context Protocol (MCP) tools and resources.
Why GoHyperrr?
GoHyperrr is built on three core pillars:
- AI-Observable Runtime: Built-in context tracking and observability let LLMs and AI agents query system status, trace event listeners, and audit dynamic DAG workflows autonomously.
- Pluggable Event-Native Architecture: Modules communicate asynchronously over an in-memory event bus or high-throughput NATS clusters, preventing service-level lock deadlocks.
- Machine-Mediated Transactions (AP2): Seamless verification of client-side credentials using SD-JWT verifiable claims and agent assertions to authorize secure payments.
Getting Started
To explore the repositories, check out the core repositories under our GitHub organization:
- Hyperrr Gateway: The main orchestrator, hosting the GraphQL interface and MCP SSE gateway.
- Module Development Kit (MDK): The Go SDK providing base interfaces, runtime testing frameworks, and top-level workflow engines.
- Commerce Modules: A collection of fully decoupled sub-modules (
product,cart,order,taxonomy, etc.).
Stay tuned for more updates as we release guides on building custom agentic checkout integrations!