Skip to content
badbots.news
Misbehavior2 sourcesFirst seen
Picture from the promptarmor.com article
Image: promptarmor.com

Snowflake's Cortex Agent Reads a README, Obeys a Stranger, Runs Malware Anyway

Ourtake

Allow-lists kept the agent from running most things, but not the one command that mattered.

What happened

PromptArmor reported a prompt injection attack chain in Snowflake's Cortex Agent. A hidden instruction in a GitHub README led the agent to run a command that downloads and executes remote code. Snowflake has since fixed the issue.

Sources