we try not to have too many opinions in how agents should work
every model is intensely trained to use certain tools and workflows so we try to match that environment perfectly
you can think you came up with something better but it will just underperform
a skill people at Anthropic have been using a lot recently: ELI5
/eli5 <what you want explained>
"explain like I'm someone who knows nothing about this topic, using a HTML artifact with big pictures and few words"
This week we read research from a team of academics that ran a software task across 7 agents and 5 models.
They found that in domains with a mature CLI ecosystem, agents without MCP baked in completed the task just as reliably and were 5-28x cheaper.
Full arXiv paper below
@jmklex@teortaxesTex@leo_linsky GLM just silently truncates the output when you hit porn or political content, which matches what I've seen in practice. Kimi/MiMo/MiniMax will straight up throw a content_filter error.
Pi 在 Harness Engineering 火爆半年后重新解释 Harness。
Agent = Model × Harness 早就不是新东西了。但我觉得 Earendil 想强调的是:未来真正属于用户的,不是 Model,而是 Harness。模型会不断更替进化,但你的 tools、skills、extensions、知识库等才是长期积累的复利资产。
Pi 它不会试图绑定一个最强模型,但会构建一个模型无关的个人 Harness。
可以看出未来 AI 竞争包括企业内的 AI Transformation 的关键问题会从「你用哪个模型?你用哪个 Harness?」变成「你的 AI 复利资产积累了些什么?」。
A harness turns a model into an agent. At it’s core it provides 4 things:
- a system prompt
- tools
- an agentic loop
- a translation layer across models
New blog post from Earendil co-founder @colindaymond on what a harness is, and how you can own yours. Full post below