AgentsMesh
A self-hosted control layer for running many AI coding agents across a fleet of a user's own machines from one console, scheduling and isolating each agent in its own sandboxed workspace and restarting ones that stall.
AgentsMesh is a control layer for operating many AI coding agents at once instead of one at a time. Self-hosted runner processes installed across any number of machines advertise their own capacity, and agents are scheduled onto a chosen runner or an available one from the pool, so code never leaves the operator's own infrastructure.
Each agent runs in its own pod with a dedicated Git worktree sandbox, private credentials and its own branch, so concurrent agents cannot step on each other's state. Web, desktop and iOS clients, all built on the same Rust core, give one paginated console with a multi-pane workspace and real-time terminal streaming, so a single operator can watch many agents at once.
An autopilot feature watches a pod and sends its next instruction as soon as it goes idle, within iteration caps, keeping a decision history and a way for a human to step back in.