Zenity 称单个提示可劫持 AWS 账户内所有 AgentCore 代理
A single prompt was enough to hijack every AI agent in an AWS account, Zenity researchers found
评分 66
导读
Zenity Labs 研究人员称，Amazon Bedrock AgentCore 上一个公开可访问的 AI agent 可被单个提示利用，接管同一 AWS 账户和区域内的所有 AgentCore agent。攻击利用了 AWS 内部临时云凭证接口，AWS 已修补该问题并收紧 agent 默认权限。
推荐理由
AWS Bedrock AgentCore 被指单个提示即可接管同账户代理，Zenity 披露攻击路径，AWS 已修补并收紧默认权限。
主体与标签
AWSZenityBedrock AgentCore#安全/对齐#Agent#部署/工程#行业动态
正文
Zenity Labs researchers say a single publicly accessible AI agent on Amazon's Bedrock AgentCore was enough to take over every AgentCore agent in the same AWS account and region. The attack exploited an internal AWS interface for temporary cloud credentials that agents could reach without restriction. AWS has since patched the issue and significantly tightened the agents' default permissions.
The article A single prompt was enough to hijack every AI agent in an AWS account, Zenity researchers found appeared first on The Decoder .