Breaking Claude Code: Hijacking System via Website
wunderwuzzi23 · x · 2026-09-01
A security researcher demonstrates an attack chain against Claude Code Opus 5 that hijacks the system via a website using indirect prompt injection. The writeup details the exploit chain and emphasizes that security invariants are mandatory.
More from Safety
- Dev discusses training models to ignore external instructions in tool calls — williawa · 2026-09-01
- Ex-Meta AI Security Head Challenges Default Thinking on AI-Related Hacking Incidents — drhyrum · 2026-09-01
- Call for OpenAI to release 70k+ message board logs — scaling01 · 2026-09-01
- METR post seen as plea for lab nationalization amid AI takeover debate — nptacek · 2026-09-01
- LLMs shouldn't run unsupervised, verify every generation — gerardsans · 2026-09-01
- Security researcher mocks 'AI will be undetectable when rogue' claims — nptacek · 2026-09-01