Anthropic launches free AI-powered OSS Scanner, finds 29,000+ potential open-source vulnerabilities
mark_k · x · 2026-10-09
Anthropic has launched OSS Scanner, a free AI-powered vulnerability scanning service for open-source projects. It uses Anthropic's strongest models, including Claude Mythos, to automatically scan code for security vulnerabilities and produce detailed reports with reproductions and proposed fixes.
Key figures: over 29,000 potential vulnerabilities discovered so far; in one validation test, 85 of 97 high-severity findings met Anthropic's disclosure standards.
The pitch: most open-source projects lack resources for security audits, and giving maintainers continuous access to frontier models for scanning could make the entire software ecosystem dramatically more secure.
More from coding & agent
- huashu-art-motion: Open-Source Agent Skill Animates 35 Art Styles With Code, 2.8k Stars — AlchainHust · 2026-10-09
- Skillry Adds 9 Languages, Shares the Agent Skill Behind Its Localization Video — yihui_indie · 2026-10-09
- Karpathy: 'Prompting is going away' — Graphs are the end state of the LLM → Agents evolution — VeryWellVersed · 2026-10-09
- Cerbere-AG: A control layer for monitoring and gating AI agent tool calls — freakingmus · 2026-10-09
- When agents edit the wrong paragraph: handling duplicate anchors in text replacement — memokris · 2026-10-09
- AgenticBBO-Bench benchmarks LLM agents for black-box optimization; GPT-6 Astra and DeepSeek-V4.1-Flash on Pareto frontier — Ming Chen · 2026-10-09