John Lin

Skills + CLI and MCP may be used differently depending on the scenario. Using Contex7 as an example please share how you use them.

github.com/DEVtheOPS/open… Compared to Claude Code, OpenCode does not yet natively support OpenTelemetry for monitoring. However, with the OTEL plugin, metrics and logs can still be exported to an OpenTelemetry Collector.

@Context7AI Does it mean Context7 MCP will be deprecated?

Introducing Context7 CLI! 🎉 The modern, token-friendly way to get up-to-date documentation for AI. 90-second demo 👇

@plateau 過去因為寫 Makefile 難一些，所以release還是得靠一些手動進行，現在我都會跟Ai説幫我做 一個release 的target 然後讓Ai使用makefile自動化 未來我也能持久化整個流程

雖然我的 github release 流程沒有完整的串起來自動化，但現在 claude code 已經可以完全照著我之前做的方式，在 local 把該做的事做一做，透過 gh 來 release 了。也就是說: 現在連 release 也是說一句 "幫我 release 吧" 就完事。 github.com/plateaukao/ein…

@plateau @Evan_Lin 最近有看到類似的應用 voxtype.io

今晚一兩個小時的成果 (結果還花不到一天 15% 的 quota): github.com/plateaukao/whi…

Then, It seems there’s a power outage! 😂

Finally a sunny day in 賢島!

@manj10 Thanks for everything! it’s been great working with you.

LINEヤフーを退職します manj.io/blog/2026/reti…

@dhh @FireworksAI_HQ Why choose FirewallsAI instead of OpenCode Zen with Kimi K2.5? Is it actually better?

Kimi K2.5 at 200tps via @FireworksAI_HQ is so good at this kind of basic "I just need help with awk or sed syntax" Linux work that I often just two a pair of twins like this.

目前常使用的 LLM 為 Kimi K2.5 模型 host 在美國，體感非常快

訂閱方案價格很有競爭力

開始轉用opencode了 目標替換claude code 的訂閱計劃

Pretty draconian from Google. Be careful out there if you use Antigravity. I guess I'll remove support. Even Anthropic pings me and is nice about issues. Google just... bans? news.ycombinator.com/item?id=471158…

everyone's talking about their teams like they were at the peak of efficiency and bottlenecked by ability to produce code here's what things actually look like - your org rarely has good ideas. ideas being expensive to implement was actually helping - majority of workers have no reason to be super motivated, they want to do their 9-5 and get back to their life - they're not using AI to be 10x more effective they're using it to churn out their tasks with less energy spend - the 2 people on your team that actually tried are now flattened by the slop code everyone is producing, they will quit soon - even when you produce work faster you're still bottlenecked by bureaucracy and the dozen other realities of shipping something real - your CFO is like what do you mean each engineer now costs $2000 extra per month in LLM bills

Typeless iOS 支援將中文口述翻譯成任何語言

LINE channel 出了！

LLM輔助開發會迅速累積技術債，真正需要的新能力是把節奏拉回來，並把驗證能力提升到能跟得上生成速度。

「測試不是萬靈丹」，因為測不到可維護性與可讀性，且不理解就很難有效 code review。

AI 工具的價值需要專業者來駕馭，而不是讓沒有開發背景的人利用 AI 取得「錯誤的自信」