Anthropic 新发的这篇多 Agent Harness 实践值得一读 借鉴 GAN(生成对抗网络)的思路,把
Anthropic 新发的这篇多 Agent Harness 实践值得一读
借鉴 GAN(生成对抗网络)的思路,把 agent 拆成 Generator + Evaluator,解决两个关键问题:
1️⃣上下文焦虑:模型在长任务中越来越差,接近 context 上限时会匆忙收工。解法不是 compaction(压缩对话),而是 context
Anthropic @AnthropicAI New on the Anthropic Engineering Blog:
How we use a multi-agent harness to push Claude further in frontend design and long-running autonomous software engineering.
Read more: