← Back to the boardNews
Devs Hit by Trojanized Claude Code Extension for Cursor
Why it’s here: A trojanized AI coding extension and a browser sandbox escape in Cursor are specialist cybersecurity findings bearing on AI agent security and supply-chain risk, relevant to cybersecurity policy even without an explicit US policy hook.
Trace across the boardItem
Oct 30Dark ReadingDevs Hit by Trojanized Claude Code Extension for CursorAnthropic's Claude Code extension on the Cursor marketplace was trojanized on Oct. 29, 2026, delivering a backdoored version 2.1.313 that stole data and could grant persistent SSH access.1 outletCyber9 linked
Connected
9 links- NewsCursor confirms security breach led to malicious Claude Code extensionBoth involve Anthropic, Claude Code, Cursor
- NewsAttackers Crash Claude's 'Office Party' Flaw to Run Malicious Code on LinuxBoth involve Anthropic, Claude Code
- NewsMeasuring AI agent autonomy in practiceBoth involve Anthropic, Claude Code
- NewsDetecting and countering misuse of AI: August 2025Both involve Anthropic, Claude Code
- NewsWhen AI builds itselfBoth involve Anthropic, Claude Code
- NewsTrustworthy agents in practiceBoth involve Anthropic, Claude Code
- NewsAnthropic Economic Index report: Learning curvesBoth involve Anthropic, Claude Code
- NewsEnabling independent research on how people use ClaudeBoth involve Anthropic, Claude Code
- NewsAI Coding Agents Leak Thousands of Secrets to GitHubBoth involve Claude Code
About
- Anthropic
- Claude Code
- Cursor