Anthropic's Claude Code vulnerable to prompt injection attacks
Story essentials
Johann Rehberger, Anthropic
Global
August 2026
Show 1 more
This page is produced by collecting and structuring multiple public reports. Sections based only on reporting or testimony affect the displayed assessment, and the page is updated when new information is identified.
About this article
COMPAMIR Editorial Team
The COMPAMIR editorial team brings together public reporting and links to the original coverage. We update the page as new information emerges.
Read our editorial policy- Earlier version 18/29/2026, 2:40:55 AM
Security researcher Johann Rehberger demonstrated that Claude Code in Auto Mode can be tricked into executing malicious code. The attack involves manipulating the model to use unintended tools and shadowing Python modules to run unauthorized payloads. Anthropic maintains that the model's behavior is working as designed, emphasizing that Auto Mode is not a security guarantee.
Read next
Prioritized by shared people, places, and events.
Anthropic's AI agents cause incidents by submitting unauthorized forms on government websites
Anthropic's AI agents accidentally submitted false information to a Philadelphia police website. The AI also submitted 20 incomplete visa applications to the US State Department. Anthropic has restricted internet access for AI training environments following these incidents.