aiqbench takeAI Safety

This report highlights how cybercriminals are adapting existing infostealers to harvest sensitive local data—such as API tokens and prompt histories—from AI coding agents like Cursor and Claude on compromised endpoints. The threat stems from the predictable storage locations of these tools rather than a vulnerability within the AI models themselves, signaling that agent adoption expands the attack surface for credential theft without requiring new exploits.

Sources coveringCybersecurity News, Fortune

What this isA model's reading of the article and related coverage, written for aiqbench. Not the publisher's words.

Hackers Target Claude, Cursor and Codex AI Agents to Steal Tokens and Prompt Histories

Cybersecurity NewsSeptember 9, 20264 min read

Cybercriminals are widening the reach of information-stealing malware by targeting the local data created by AI coding agents. The shift puts access tokens, saved connections, prompt histories, and project records at risk on already infected computers. The activity does not point to a newly discovered flaw in Claude, Cursor, or Codex. Instead, it shows criminals […] The post Hackers Target Claude, Cursor and Codex AI Agents to Steal Tokens and Prompt Histories appeared first on Cyber Security News.

Cybercriminals are widening the reach of information-stealing malware by targeting the local data created by AI coding agents. The shift puts access tokens, saved connections, prompt histories, and project records at risk on already infected computers. The activity does not point to a newly discovered flaw in Claude, Cursor, or Codex. Instead, it shows criminals adapting established stealers to collect valuable files stored in predictable local folders, a concern echoed in…

Excerpt shown under the publisher's RSS terms · 71 words · full text and figures are on the publisher's site

Read at Cybersecurity News ↗Ask about this article