TLDR Tech / Launches
Anthropic offers free AI-powered scanning for open-source security
Anthropic has introduced the OSS Scanner, a complimentary tool that employs AI models like Claude Mythos to automatically detect and report potential vulnerabilities in open-source software. Over six months of testing, the scanner identified over 29,000 potential vulnerabilities, forwarding numerous unverified reports and suggested fixes to maintainers. A review by expert penetration testers found an 88% validity rate for critical and high-severity findings, with only one false positive.
Read the full story