How do I turn on dynamic workflows in Claude?

Select the "multiagent_20261001" agent type, or run "/claude-api managed-agents-onboard" in Claude Code. Anthropic's documentation explains the setup.

How much better is the multi-agent workflow at finding bugs?

In a 116,000-line codebase with 70 hidden bugs, a single agent caught between 14 and 27 per run. The dynamic workflow consistently caught 66.

Large Language ModelsAI Coding AssistantsTHE DECODER2m agoOct 10, 2026, 04:00 JST

Anthropic's Claude Managed Agents orchestrates 1,000 parallel agents

Send on LINE

3 Key Points

What happened
Anthropic added dynamic workflows to Claude Managed Agents. A lead agent plans, farms tasks out to up to 1,000 parallel sub-agents, then merges the results. In a 116,000-line codebase with 70 hidden bugs, one agent caught 14–27 per run; the workflow consistently hit 66.
Why it matters
The 66-of-70 result suggests parallel sub-agents can find substantially more hidden bugs than a single run, though Anthropic notes these workflows burn through a lot of tokens and recommends starting small.

WHO IT HITSSoftware engineering teams evaluating automated code review or bug-hunting tools will weigh the higher bug-detection rate against token costs, since Anthropic itself recommends starting small.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, in your inbox every morning.

Get news like this every morning, free

Context & Analysis

Anthropic's Claude Managed Agents have offered managed agent infrastructure for some time. What is new is dynamic workflows: a lead agent creates a plan, distributes tasks to sub-agents, and merges the results when they are done, with up to 1,000 agents able to run in parallel per execution.

The company's own testing compared this setup to a single agent. In a 116,000-line codebase with 70 planted bugs, one agent caught between 14 and 27 per run, while the dynamic workflow consistently reached 66. Anthropic cautions that these workflows can burn through a lot of tokens and recommends starting small.

Whether the bug-detection gains hold across different task types remains to be seen — Anthropic advises testing with your own workloads. For teams weighing the approach, the trade-off appears to be token spend against broader bug coverage, and the company's own guidance points toward small-scale trials first.

FAQ

How do I turn on dynamic workflows in Claude?: Select the "multiagent_20261001" agent type, or run "/claude-api managed-agents-onboard" in Claude Code. Anthropic's documentation explains the setup.
How much better is the multi-agent workflow at finding bugs?: In a 116,000-line codebase with 70 hidden bugs, a single agent caught between 14 and 27 per run. The dynamic workflow consistently caught 66.

THE DECODERRead Original Article

Send on LINE

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Get it free

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.