The Agent in the Middle: How AI-Mediated Access Reshapes Web Vulnerability Exploitation

Summary

The researchers introduce VELA, an executable benchmark of 44 vulnerabilities across twelve Web application families, with agent-facing interfaces derived from legitimate workflows. They evaluate three commercial agents and find that interface design, not agent choice, mainly sets exploitability: 91% of vulnerabilities stay exploitable under the reference design, and retention moves from 39% to 95% across interface configurations versus at most 4.5 points across agents. Some restrictive interfaces also block legitimate workflows.

Related items

• HighCVE-2026-115924: Rembg unauthenticated file disclosure through public /api/remove endpointSimilar attack · NVD/CVE Database
• HighCVE-2026-115915: vLLM image loading flaw allows EPS payload dispatchSimilar attack · NVD/CVE Database
• LowThe RealReal won’t let AI agents hog designer clothingSimilar attack · The Verge (AI)
• HighCVE-2026-115912: vLLM arbitrary Python execution via forged shared-memory multimodal dataSimilar attack · NVD/CVE Database
• MediumOpenAI, Anthropic disclosures spotlight growing AI containment challengeSimilar attack · CSO Online