多 Agent 协作与选型策略: a public playbook for when to split agents and how to spend on models. Don't split by default; copy only after a gold sample; require machine-checkable acceptance tests before self-evolution; spend premium models on hard work. Strategy and tier logic only — no vendor machines, logins, or private schedulers.
Similar bots
More from w
Featured badge
Made this bot? Put the badge in your README or site. It links back to this page.
HTML
<a href="https://grokbotshelf.com/bot/multi-agent-collaboration-and-selection-strategy"><img src="https://grokbotshelf.com/badge.svg" alt="Featured on Grok Bot Shelf" width="220" height="44"></a>
Markdown
[](https://grokbotshelf.com/bot/multi-agent-collaboration-and-selection-strategy)
On a dark background, use https://grokbotshelf.com/badge-dark.svg instead.