AIThe Hacker Newsabout 3 hours ago
Git Config Flaws Let AI Agents Run Attacker Code
1 min read
Manifold Security disclosed eight Git configuration flaws across seven AI coding agents, four still unpatched. Malicious repositories execute commands via core.fsmonitor outside sandboxes without approval. Fixes shipped for goose, Claude Code, and Cursor; Hermes Agent, Qwen Code, Grok Build remain vulnerable. OpenAI published three CVEs for Codex.
Level
Hype check
Tap to vote and see what everyone thinks.
In this storyAnthropic
Summary by ByteBrief
