Anthropic's open-source framework for AI-powered vulnerability discovery
Summary
The article describes Anthropic's open-source Defending Code Reference Harness, an AI-powered vulnerability discovery and remediation pipeline. It outlines the seven-stage process (recon, find, verify, dedupe, report, patch) and emphasizes safety, sandboxing, and that this is a reference implementation rather than a product, with a managed option available via Claude Security. It serves as a resource for building automated security testing pipelines using AI, with guidance on porting and customizing.