Google Deletes 3 ADK AI Workflows After Malicious GitHub Issue Could Trigger Privileged Agent
Researchers discover a prompt-injection vulnerability that allowed public users to trigger privileged code-fixing tasks.
Google deleted three AI agent workflows from its Agent Development Kit (ADK) Python repository. Pillar Security showed that a public GitHub issue could manipulate a triage agent into triggering a privileged code-fixing agent. The researchers said the public agent could be prompt-...
The Hacker News
Source

Generated insight by OnAirToday AI
This incident highlights the dangerous escalation potential when autonomous agents are granted excessive repository permissions without human oversight. It underscores the critical need for strict authorization boundaries to prevent public interactions from triggering privileged administrative actions.
Comments
Keep comments relevant and respectful. No links, advertising, self-promotion, or spam.