Lab: Exploiting AI agents to exfiltrate sensitive information
Summary: PortSwigger Research published a lab in which an AI-powered site scanner holds sensitive user data, including API keys, and has been given the login credentials for the user carlos so it can explore authenticated areas of the site. The lab is vulnerable to indirect prompt injection, and solving it requires manipulating the LLM's behavior through third-party content to exfiltrate and submit carlos's API key.
Bottom line: The scanner was handed carlos's credentials and access to API keys as part of its normal job, so third-party content it reads during a scan becomes a path to the credentials and keys it was trusted with.
Read the original at PortSwigger Research ↗
- Topics
- AI Security
- Related
- SECURITY AFFAIRS AI-CYBERSECURITY NEWSLETTER ROUND 8 Cyber Risks · Nov 15, 2026Pentagon Renames CDAO ‘Super Intelligence Office’ Cyber Risks · Nov 15, 2026186 Fake AI Domains Redirect Users to Click Fix Lures, Browser Malware, and Search Hijackers Cyber Risks · Nov 12, 2026