Anthropic lets Claude coordinate up to 1,000 AI agents in one run

Anthropic has added dynamic workflows to Claude Managed Agents. A single run can now coordinate up to 1,000 AI agents working in parallel on big, complex tasks. Here is how it works: a main agent plans the job, splits it into smaller pieces, and hands them out to sub-agents. When they are done, the main agent pulls all the results together. This works well for large jobs like code reviews. To show this off, Anthropic hid 70 bugs in a codebase with 116,000 lines of code. A single agent caught 14 to 27 bugs per run, while the dynamic workflow consistently found 66.