Anthropic has quietly expanded its agentic coding tool Claude Code with a session mode called Ultracode, which fans out dozens or even hundreds of parallel sub-agents across a single task and has each one cross-check its own findings before delivering one thorough, verified result. The feature arrived alongside the release of Claude Opus 4.8 and is aimed at complex engineering work that overwhelms conventional single-pass assistants.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.