Malicious Jet Brains Plugins Steal AI API Keys as Chrome Extensions Capture Chatbot Chats
Summary: Aikido Security researchers found 15 malicious plugins on the JetBrains Marketplace, active since late October 2025 and published as recently as June 10, 2026, that pose as DeepSeek-based AI coding assistants and work as advertised while sending the user's AI provider API key in plaintext over HTTP to an attacker-controlled server at 39.107.60[.]51. Two plugins, CodeGPT AI Assistant and DeepSeek AI Assist, each show more than 25,000 downloads, and JetBrains removed all 15 plugins on June 17, 2026, blocked seven publisher accounts, and disabled the plugins in installed IDEs. Separately, researcher Jean-Marie R. codenamed PromptSnatcher two Chrome ad blocker extensions that intercept full AI conversation history, model usage, and subscription tier from eight platforms including ChatGPT, Claude, and Gemini, and send it to operator-controlled infrastructure.
What's compelling: StepSecurity's June 19, 2026 check found the attacker's command-and-control server still live and answering API requests three days after JetBrains pulled the plugins, so any stolen key not yet rotated can still be used, and the operator collects payment from plugin users while the genuine key owners pay the bill.
Read the original at The Hacker News ↗
- Topics
- AI Security
- Related
- SECURITY AFFAIRS AI-CYBERSECURITY NEWSLETTER ROUND 8 Cyber Risks · Nov 15, 2026Pentagon Renames CDAO ‘Super Intelligence Office’ Cyber Risks · Nov 15, 2026186 Fake AI Domains Redirect Users to Click Fix Lures, Browser Malware, and Search Hijackers Cyber Risks · Nov 12, 2026