ITSMBench
IT service management environment for testing agents inside a system of record
An open-source IT service management benchmark/environment in the Enterprise Worlds suite for evaluating AI agents on realistic multi-turn, policy-governed enterprise workflows with persistent state and deterministic final-state grading.
Recent stories
A Renmin survey, Cline's SDK notes, and AutomationBench trajectory reviews point beyond context windows to harness design and verifier quality. A failure-mode paper maps issues across models, memory, tools, users, and environment.
Enterprise Worlds opened ITSMBench for executable enterprise-agent tasks with persistent state, simulated users, 93 tools, and deterministic grading. The benchmark starts with IT service-management workflows.