https://left4more.com/tags/ai-security/Recent content in Ai-Security on Left 4 MoreHugoen-auSat, 02 May 2026 10:00:02 +1000https://left4more.com/posts/cursors-security-mess-claudes-new-effort-levels-an/Sat, 02 May 2026 10:00:02 +1000https://left4more.com/posts/cursors-security-mess-claudes-new-effort-levels-an/<p>April was a big month. Possibly too big. Between a critical RCE in Cursor, Anthropic shipping Opus 4.7 with three silent breaking changes, and the &ldquo;ultra prefix&rdquo; commercial model crystallising into something real, there&rsquo;s a lot to unpack. I&rsquo;m going to focus on the three things I can&rsquo;t stop thinking about.</p> <hr> <h2 id="the-cursor-cve-should-have-been-front-page-news">The Cursor CVE Should Have Been Front-Page News</h2> <p>Let&rsquo;s start here, because this one genuinely alarmed me.</p> <p>CVE-2026-26268 is a CVSS 9.9 remote code execution vulnerability in Cursor versions prior to 2.5. The mechanism is nasty: a malicious actor embeds a bare repository inside a legitimate-looking public repo, with a crafted pre-commit hook. When the Cursor agent runs a <code>git checkout</code> as part of a routine task — something agents do constantly — that hook fires automatically. No warning, no confirmation prompt, nothing. You just handed someone a shell.</p>