Claude Code · Anthropic · Claude · Apple · The Register
Researcher shows how Claude Code can be tricked simply by asking it to summarize a website
Compiled by KHAO Editorial — aggregated from 1 source. See llms.txt for citation guidance.
◌ Single Source
More prompt-injection hijinks from wunderwuzzi.
Key facts
- Across three variants tested five times each, which Rehberger noted were small samples, he reported success rates between 60 percent and 80 percent
- Anthropic’s Claude Code running Opus 5 in Auto Mode can be tricked into executing attacker-controlled code simply by asking the coding agent to summarize a website
- The website returns a 303 response, and redirects to a malicious ZIP archive, which Claude then downloads
- In a blog and video demo, he detailed how to hijack Opus 5 in Auto Mode, which is the default setting for Claude as of mid-August
Summary
Security Security vets rally around $4 paper password books for sale in Australia. Ai and ML Slop factory bans Russians for using slop factory to create slop. DATABASES AWS buys DuckLabs, the people behind the popular in-process OLAP database. Anthropic’s Claude Code running Opus 5 in Auto Mode can be tricked into executing attacker-controlled code simply by asking the coding agent to summarize a website. In a blog and video demo, he detailed how to hijack Opus 5 in Auto Mode, which is the default setting for Claude as of mid-August.