dsh-agent-hub

Coordinating multiple AI agents on real projects often means chasing scattered outputs, opaque states, and hard-to-reproduce failures. dsh-agent-hub collapses the entire workflow into a single group-chat room: a requirement comes in, a verified deliverable comes out, and every step of decomposition, dispatch, execution, acceptance, and knowledge capture is visible and traceable in the message stream—no external services, no manual backend setup required out of the box.

At the centre of the system sits a coordinator brain that you assign to any member. It drives each round through a strict decision-dispatch-collect-accept-memorise state machine, and every round must explicitly answer four questions—goal, current state, available members, definition of done—before emitting structured JSON instructions. Tasks are split into atomic units where one member owns one deliverable plus one acceptance criterion, and independent tasks run in parallel. Members connect through any OpenAI-compatible API, so DeepSeek, Tongyi, Kimi, and GPT can coexist in the same team. After delivery, the brain checks every acceptance criterion one by one, issues a precise point-by-point rework list on failure, and writes passing results into a persistent group memory that carries across rounds.

The plugin ships with a self-contained backend, has zero npm runtime dependencies, and is ready the moment it is installed. It is built for developers and engineering teams who need multiple agents to collaborate end-to-end: scaffolding a feature from scratch, comparing and implementing technical options, reviewing and fixing code, or running parallel research and consolidating it into a final document—each scenario closes the loop from requirement to verified delivery inside a single chat window.