Network troubleshooting agents invent root causes when nothing is broken.

FaulT-Bench runs 200 scenarios across eight topologies, five reimplemented from public practitioner labs, spanning genuine faults, false fault reports, wrong device attribution and wrong root-cause claims. SADE, ReAct and Claude Code are all near-saturated on accurate tickets and robust to misdirection, then degrade sharply when the network is healthy and the ticket is wrong, probing until a benign condition can be promoted to a root cause. Rewriting 72 false-premise tickets into five reporter personas shows wording matters more than content: a confidently wrong report is handled about as well as an accurate one, while a vague underspecified one collapses performance. (arXiv)