Security

Claude Code 2.1.222 fixes two agent-isolation failures, yet the npm stable tag still resolves to 2.1.220. Check which build your own team is actually on.

How the June 2026 Miasma npm worm beat standard defenses — plus a 7-step checklist to harden your AI agent pipeline against supply chain attacks.

Claude Opus 4.8 and Sonnet 4.8 are unconfirmed. The real buyer signal is a tiered Anthropic roadmap built around routing, cost and security governance.

Robin Ebers put a useful phrase on a problem every AI software team eventually meets: the dangerous bug is not always the one an agent misses. Sometimes it is the one an agent quietly routes around. That does not mean

Claude Code v2.1.113 ships native binaries for the first time, adding sandbox.network.deniedDomains for binary-level security. What this means for developers.

How Anthropic's new security tool finds bugs that static analysis misses — and how to use it effectively.