Agent: Claude Code, Claude SecurityLLM: Claude 3.5, Claude Mythos#security#vulnerability-detection#code-scanning#autonomous-agent#cybersecurity
A reference implementation from Anthropic for finding and patching vulnerabilities autonomously using Claude. Includes interactive skills for threat modeling, scanning, and patching plus an autonomous harness for C/C++ memory vulnerabilities.
