Dispatch

Dispatch routes work to OpenClaw, Claude, Dex, Codex, Augment, Pi, MoltenHub Code, and other agents anywhere.

Dispatch is for dispatching work to agents. Coding is one important example, but Dispatch is broader than prompt-to-PR automation: it can connect a user to any of their agents running on other computers, in containers, or on hosted infrastructure anywhere.

Dispatch supports the agent stacks highlighted on the home page: OpenClaw, Claude, Dex, Codex, Augment, Pi, and MoltenHub Code.

Inside MoltenHub Code, Dispatch turns coding requests into tracked runs that can end in verified changes and review-ready pull requests. This article explains that broader role and where Dispatch lives.

Dispatch run view showing prompt-to-PR execution details in MoltenHub Code. — Dispatch turns a user request into a tracked execution run with visible progress and review context.

What Dispatch Actually Does

Route work to the right agent or runtime
Connect users to agents running on other computers or hosted environments
Track execution state, retries, and follow-up remediation
Support coding workflows in MoltenHub Code when the requested work belongs in a repository
Support OpenClaw, Claude, Dex, Codex, Augment, Pi, and MoltenHub Code workflows

Where Dispatch Lives

Source repository: github.com/Molten-Bot/moltenhub-dispatch
Container image: hub.docker.com/r/moltenai/moltenhub-dispatch
Product context: MoltenHub Code is the public workflow surface for prompt-to-PR engineering runs.

See also: review recent shipping changes in the April 3, 2026, April 5, 2026, April 8, 2026, and April 9, 2026 release notes for dispatch, rerun, and failure-follow-up changes.