Anthropic's Biggest AI Coding Projects Verified With Test Suites, Not Other Agents
Anthropic shipped dynamic workflows in Claude Code on May 28, letting Claude write a script that coordinates up to 1,000 agents. This walkthrough builds a document audit you can point at a draft, a filing, or a set of marketing claims, with the rubric that decides what counts as proof. Then it runs that audit on this article: 204 agents, 189 claims, nine minutes. It caught two errors two earlier fact-checks missed, and raised fifty false alarms doing it.
How to Audit a Document With a Team of Claude Agents
Implicator PRO Briefing / 21 Jul 2026
Pro Members Only
On May 28, Anthropic shipped a feature that lets Claude write its own orchestration script and run up to 1,000 agents from it. This walkthrough builds one useful thing with it: a document audit that splits a draft, pulls out every checkable claim, and hands each one to a separate agent that must quote the passage supporting it or mark it unsupported. Then it runs that audit on this article. 204 agents, 189 claims, nine minutes, two real errors caught that two earlier fact-checks had missed, and about fifty false alarms. The harder question sits underneath: what Anthropic's own largest agent projects actually trusted to tell them their work was correct.
New to Implicator PRO? Subscribe for $8/month — a new deep dive every Tuesday morning 3am PST.
Editor-in-Chief and founder of Implicator.ai. Former ARD correspondent and senior broadcast journalist with 10+ years covering tech. Writes daily briefings on policy and market developments. Based in San Francisco.
E-mail: editor@implicator.ai
The Morning Briefing
Get the Morning Briefing in your inbox.
Sign up to our free daily morning newsletter and free member articles. Only our special weekly Pro Briefing is available for $8/month.
IMPLICATOR.ai
