Agensh is a multi-agent harness that removes the central-orchestrator bottleneck limiting existing multi-agent frameworks by having concurrent workers self-organize — claiming sub-tasks, sharing findings, verifying results, and merging progress asynchronously through a shared workspace, message interface, and shared context. Tested on the hardest ProgramBench tasks, scaling from 1 to 1,024 agents raised the final test-pass rate on one task from 33.89% to 55.06%. The authors present agent count as a new, largely unexplored scaling dimension for multi-agent systems operating under hard latency budgets.
