Automate Your Workflows with Kimi Work - Tutorial #3 See how Kimi fits into your daily workflows: - Work across your local projects, browser, and computer. - Break down complex tasks using Swarm; iterate in Goal mode. - Customize how Kimi works with Plugins, AGENTS.md and Dashboard. What should... we cover next? Let us know in the comments!show more

Sooo ... Let's talk about 3.1

Everybody's waiting for 3.1

where did badge go team?

我就希望额度能透明点

Wen Kimi Agent App 😮?

I'm waiting, k3.1, kimi's aesthetics is really good.

快点发新版本啊，急死我了

Kimi is making AI-powered workflows more practical and efficient! Would love to see a tutorial on building fully automated workflows from scratch.

We need Kimi Bot

workflow automation across local files is the useful part is kimi work free or paid only ?

You should cover Kimi 3.1 next thanks

I have a wild idea: (Digital) BaaS [Business As A Service] Agents start a business platform to create, run and manage (digital) businesses. From E-Commerce to Digital marketing agencies.

为啥麦克风的音量这么小？不应该。专业制作不能丢失这个水准。

Is Kimi Code able to control Brower and do all the same things? My only computer is Linux based so im not able to use Kimi Work

Gotcha from testing kimi-k2.7-code through OpenRouter today: with a tiny max_tokens, the reasoning token used up the budget. Result was HTTP 200, finish=length, empty content, still billed. If you script it, leave headroom for reasoning tokens.

the workflow tutorials are cool, but everyone’s really waiting for k3.1 😭

快发新模型！！

The Swarm-plus-Goal pattern is the planner/worker split done right: decompose once, then iterate per piece. Splitting across local projects, browser and computer is the harder half — different tools, one goal state.

This tutorial is super helpful. Swarm and Goal mode sound powerful

Nice work. Can you show how to use Swarm on a small task? A simple demo would be great.

想看任务跑到一半出错了怎么接着跑

Kimi 3.1 !!!

第三期把 Swarm 拆任务 + Goal 模式迭代讲得挺清楚，关键是能跨本地项目、浏览器和电脑一起干活。AGENTS.md + Plugins 自定义这块很实用，团队规则写一次就能复用。下一期想看：长任务中途失败怎么回滚、怎么看每个子任务的进度。

赶紧发新模型

A deep dive into AGENTS.md best practices with Swarm would be awesome! Curious to see how you handle complex task handoffs across larger repos.

Krass, ich hab echt Bock auf euer k3.1! 🔥

where is k3.1 btw

So ,I think you need Kimi K3.1

你们应该合并 Kimi Work 和 Kimi Code

别扯这些没用的，快点发 k3.1 和 k4 啊

Using the same AGENTS file Cursor and Codex read is a smart call, so project rules travel between tools. Question: how does Goal mode decide a task is done, tests or self-judgment? That stop condition is what I watch most when comparing it with Grok Bot on its own cloud box.

the real test is how much busywork it can take off your plate without needing constant supervision.