Docker fixes Sandboxes file access flaw
Endor Labs detailed in a company blog post a vulnerability in Docker Sandboxes that could let AI coding agents access files outside approved workspaces. Docker fixed CVE-2026-103916 in Sandboxes version 0.51.0, released November 3.
The flaw affected the sbx cp command, which copies files between a sandbox and the developer's computer. An agent could place a symbolic link inside an approved workspace, causing a later copy command run by the developer to read or overwrite files elsewhere on the host system.
Exposed files could include cloud credentials and SSH keys. An attacker could also use prompt injection to make a coding agent prepare the symbolic link. Users should upgrade to version 0.51.0.
We hope you enjoyed this article
Consider subscribing to one of our newsletters like Cybersecurity AI Weekly, AI Programming Weekly or Daily AI Brief.
Also, consider following us on social media:
More from Cybersecurity
Nov 3 Irregular Blames AI Model Test Escapes on Human Error Nov 3 OpenAI Raises Bug Bounty Rewards to $172,000 Nov 3 NCSC Says AI Will Accelerate Cyberattacks and Widen Security Gaps Nov 3 CrowdStrike buys Fiddler AI and Dreadnode in deals reported at $495 million Nov 3 CyberKord raises €2.7 million for AI security automationAI Programming Weekly
Weekly news about AI tools for software engineers, AI enabled IDE's and much more.
Industry analysis
2025 Global Business Services Agenda: Gen AI Takes Center Stage
This industry analysis by The Hackett Group explores the transformative impact of generative artificial intelligence (Gen AI) on global business services (GBS) in 2025. The study highlights the shift from exploration to acceleration of Gen AI initiatives, with 89% of executives advancing these projects to improve customer satisfaction, innovate products, and reduce costs. The report also discusses the challenges and strategies for successful Gen AI adoption, emphasizing the need for a technology-enabled operating model and the importance of reskilling the workforce.
Read moreYou may also like
x402 SDK Flaws Put AI Agent Wallets at Risk
Anthropic uses Claude to find 493 attack surface issues
OpenCode breach may have exposed stored AI provider keys
External test found gaps in OpenAI agent monitoring
Northflank Breach Leads to Unauthorized AI API Use
Daily AI Brief: the AI news that matters, in your inbox.