Graphon is a Python graph execution engine for agentic AI workflows.
The repository is still evolving, but it already contains a working execution engine, built-in workflow nodes, model runtime abstractions, integration protocols, and a runnable end-to-end example.
- Queue-based
Engineorchestration with event-driven execution - Graph parsing, validation, and fluent graph building
- Shared runtime state, variable pool, and workflow execution state
- Built-in node implementations for common workflow patterns
- DSL import support with Slim-backed LLM nodes
- HTTP, file, tool, and human-input integration protocols
- Extensible engine layers and external command channels
Follow development setup, then run the Slim LLM example. It includes both DSL import and direct Python graph construction.
- Architecture: public entry points, execution flow, and state ownership.
- Development guide: change surfaces and host integrations.
- Repository knowledge: component guides and maintenance.
- Contributing: setup, validation, and contribution workflow.
Apache-2.0. See LICENSE.