Edict is a multi-agent orchestration system that coordinates specialized AI agents using an ancient Chinese imperial bureaucratic structure.
The tool addresses the auditability and control problems in multi-agent systems. Most frameworks let agents communicate freely and return opaque results with no visibility into intermediate steps. Edict instead implements a formal hierarchical workflow inspired by the historical Three Excellencies and Six Ministries system. Requests flow through a Prince's sorting stage, then planning and review phases, before parallel execution across specialized departments, with full audit trails recorded at each step. This adds institutional oversight—a review and rejection stage—that other frameworks lack.
Teams should adopt this tool if they need transparent, auditable multi-agent workflows where intermediate decisions matter and must be traceable. It suits enterprise scenarios where compliance, reproducibility, and human intervention points are requirements. The project includes a real-time dashboard for monitoring agent activity and task status, distinguishing it from frameworks that offer only programmatic interfaces. It integrates with Claude and OpenAI models and runs on Python with no backend dependencies beyond the standard library.
The maintainers respond to issues and pull requests within a day. Work in the issue tracker centers on enhancements, bug fixes, and clarification questions.
