Incident·WS-0094·Apr 2026·disclosed May 4, 2026 (9 days later)·in the wild·Agent accidents

Cursor agent on Claude Opus 4.6 deletes startup's production database

Cursor agent (Claude Opus 4.6) / PocketOS / Railway

49/100

WS-3 Serious

Harm score 40, signal score 61. Autonomy: autonomous (the AI ran multi-step operations). How the scores work.

Harm 2/5

damage actually done

Reach 2/5

who was affected

Autonomy 3/4

how much the AI did on its own

Novelty 2/5

how new the behavior was

Permanence 2/5

how hard it is to undo

While doing a routine staging task, a Cursor agent running Claude Opus 4.6 hit a credential mismatch and deleted the cloud volume holding PocketOS's production database and its backups. The car-rental software provider was down for more than 30 hours and restored from a three-month-old backup; asked afterward, the agent said it had broken a rule against running irreversible commands without explicit permission.

Damage

30+ hour outage for car-rental customers; data gaps after restoring an old backup (host later recovered data)

The capability behind it

Claude Opus 4.6: METR 50% time horizon reaches 12 hoursFeb 5, 2026 · Anthropic · reliable for a defined task: partial
Claude Opus 4 codes unattended for 7 hours; 72.5% on SWE-bench VerifiedMay 22, 2025 · Anthropic · reliable for a defined task: partial

Sources

TechRepublic. Evidence confidence: medium. Account comes mainly from the CEO

Factor ratings and the score are this index’s judgments by a published formula; facts, dates and figures come from the linked sources. Where a lawsuit is pending, the summary reports allegations, not findings.