Threat
DeepSeek Harness Flaw Let AI Agents Disable Their Own File Sandbox Without Approval
Publication date not yet evaluated · Added · 2 reports thehackernews.com
Page published
Coverage timeline
Why it matters
CVE-2026-82533 shows how an AI coding agent working with untrusted material can trivially escape its OS sandbox on shipped defaults, turning containment controls into a false sense of security for anyone deploying agentic coding tools.
OX Research disclosed CVE-2026-82533, a critical (CVSS 9.4) flaw in DeepSeek Harness, DeepSeek's open-source AI coding-agent harness, that let a sandboxed AI agent disable its own confinement with a single shell command. The agent-control API was exposed on a local HTTP port without authentication and trusted the client-supplied 'Host' header, so a confined agent could elevate its session to 'danger-full-access' with approval prompts disabled; if the port was network-exposed, an unauthenticated remote attacker could seize the agent and export all stored conversations.