What happened
Anthropic used Claude to analyze 25 open-source projects (over 5 million lines) and found 493 issues, including 162 silent config combinations, 94 of which unintentionally disabled security features.
Why it matters
The finding suggests unnecessary attack paths can persist even in widely used code, since 11 projects all had code reachable from outside before authentication and one config contradiction stayed for 21 years.
WHO IT HITSOpen-source maintainers and security teams reviewing default configurations now have a concrete method for shrinking externally reachable code, with Anthropic's fixes already reflected in OpenSSH 10.6.
Summaries like this, in your inbox every morning.
Pick your industry and the AI tools you use, and get news related to your work every day.
Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →
Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.
Calif used GPT-5.6 Sol to uncover the critical PHP vulnerability PHP2Shell (CVE-2026-109618/CVE-2026-109619)…
The author details a setup where Claude Code reads an entry document called CLAUDE.md, writes results into rol…
Using atopile, KiCad, Freerouting and Claude Code, the developer automated circuit description, placement, rou…
The author ran 6 questions three to five times each with claude -p across 4 layouts
With gaurd.mjs typo'd in settings.json, the guard hook returned status 127 and the rm command deleted the fake…
PMDesk, a macOS app for running multiple projects with AI agents, now lets users type "@" in its terminal to o…
