AIPolicyTracker

AI incident ·

Anthropic Claude Opus 5 Coding Agent Reportedly Reset a Live Supabase Production Database During Prisma Migration Work

3 news reports Synced from source · record last edited 4 Sep 2026

In brief

An AI system built by Large language model developers, Anthropic and 1 other and deployed by Software developers, Alone_Ad_3375 (Reddit user) and 1 other allegedly harmed Software developers, Database operators and 1 other.

Risk domain
Not classified
Occurred
Coverage
3 reportsJul 2026 - Aug 2026

What happened

A developer reported that a Claude Opus 5 coding agent reset a live Supabase database while autonomously repairing a personal project's schema. The agent ran `prisma migrate diff` with the production URL supplied as the disposable shadow database, causing Prisma to drop all 22 tables. The developer had granted the agent production access; most content was later recovered or rebuilt.

Laws that address this harm

No recorded instrument yet addresses this use case where it happened. See the open queue.

Matched from the record's risk domain and country to the instruments recorded here. A reviewer can correct the match in the repository (data/external/incident_overrides.yaml).

News reports (3)

Titles link to the original publisher; report text is not reproduced here.

Who was involved

Alleged harmed party
Software developers Database operators AI agent system users

AI systems implicated

SupabaseProduction database serversPrismaLLM-integrated code assistantsLarge language modelsClaude Opus 5Claude CodeClaudeAI agent systems

Classification (MIT AI Risk Repository taxonomy)

Risk domain
—
Risk subdomain
—
Causal entity
—
Intent
—
Timing
—
Harm level
—
Sectors
—
Countries
—

Linked by editors or by text similarity in the source dataset.

Other incidents involving Software developers

Source record: incident #1676 on the AI Incident Database · all 3 reports