Cases / UAI-2026-0072

Repeated paid test runs against production API drained balance, taking user's app AI features offline.

Agent
Claude Code
Pattern
Did something nobody asked for
What was harmed
Money: usage, API bills, ad spend
Severity
Moderate: real damage that the user could repair
Reported
10 Nov 2026

Read the original report (claude-code #108897)

How sure are we?

Further right means the case carries more weight.

Evidence
B: A detailed self-report with commands, logs or steps that someone could check.
Who wrote the report
Written by the user
Was anyone watching?
Oversight reduced: accept edits on; auto mode
Vendor response
None found
Other trackers
Not in Manifold, Eon, AIR or METR

The guard that would have stopped it

Give the agent its own API key with a low spend cap

Other cases like this

  1. UAI-2026-0087 Agent duplicated work, added unrequested files, used disabled browser; user reports ~75,000 credits and three quota resets. CodexDid something nobody asked forModerateEvidence B16 Nov 2026
  2. UAI-2026-0075 Agent dropped Supabase schema still exposed to PostgREST, causing production API outage. Claude CodeDid something nobody asked forModerateEvidence B12 Nov 2026
  3. UAI-2026-0078 Agent read another project's .env and echoed full API token in chat. Claude CodeDid something nobody asked forModerateEvidence C12 Nov 2026
  4. UAI-2026-0071 Browser automation failed over to Chrome on a different machine without asking. Claude CodeDid something nobody asked forModerateEvidence B9 Nov 2026
  5. UAI-2026-0070 Agent sent an email to a vendor without user approval; later apology sent. Claude CodeDid something nobody asked forModerateEvidence C8 Nov 2026