Video wird geladen...
Video konnte nicht geladen werden
I don't understand why everyone isn't doing this yet. Anthropic's own Claude Code docs show how to run a whole team of Claudes, while Opus 5.5 only touches the plan and the merge the whole idea: agent teams in Claude Code one lead, separate teammates in their own context... show more
69,178 Aufrufe • vor 4 Tagen •via X (Twitter)
4 Kommentare

Jadu 👐vor 4 Tagen
the adversary staying review-only is the cleanest bit here lol

1Broomvor 4 Tagen
Giving the critic zero authorship is the smart call. A reviewer that didn't write a line has no ego tied up in the result, so it actually goes looking for the crack.

Martin Ronfortvor 4 Tagen
We actually went deeper on how agents can manage complex workflows like this here:

catmanvor 4 Tagen
It’s like parallel work with a code reviewer at the gates: separate ownership speeds implementation, while adversarial checks catch contract and test failures before merge.
