56kode

Frontend developer specializing in all things #React, #TS, and #JS

[56kode @56_kode]

New post 📢: how I used Claude Code to build a Chrome extension and an n8n workflow that automates my AI tech watch sharing. The whole thing in about 30 minutes.
https://t.co/JHQFDfRsfT

[56kode @56_kode]

@status_is_down I have an issue

[56kode @56_kode]

🔥 Just discovered https://t.co/mxjI22MGMS, a game-changer for Claude Code users!
It filters shell output noise before it reaches your AI:
✅ Fewer tokens = lower cost
✅ Longer sessions, no context reset needed
✅ More focused agent, less distraction
Thanks to @FlorianBRUNIAUX

[56kode @56_kode]

Why I migrated from #Cursor to #ClaudeCode (and why you should too). ⚡️

I stopped typing code and started supervising agents in the terminal. The productivity gain is real. 🤖

Read my full honest feedback here:
👉 https://t.co/tA0LD4ONVo #ai #agent #opus

[56kode @56_kode]

@kloss_xyz @karpathy @steipete @gregisenberg @rileybrown @corbin_braun @jackfriks @EXM7777 @eptwts great list, thanks

[56kode @56_kode]

@Tech_girlll Claude should post on Stack Overflow if he can't find the right answer to your problem.

[56kode @56_kode]

@dr_cintas Wow great, thanks

[56kode @56_kode]

@SleepDaizy Hi

[56kode @56_kode]

@zwebtiai Ideally, it lets the team focus on building product value instead of hunting dead code. The workflow adds a layer of safety, though manual PR review is still non-negotiable before production.

[56kode @56_kode]

Stop cleaning dead code manually! 🛑

Discover a safe workflow combining Knip to find unused exports and an AI agent to verify and remove them without breaking production. 🤖

👉 https://t.co/U8XWZZC7HC

#TypeScript #ts #ai #cursor #claudecode

[Matt Pocock @mattpocockuk]

Bad AGENTS​.md files can make your coding agent worse and cost you tokens.

Here's a prompt you can use to clean them up - PLUS a full guide for folks wanting to learn more.

Enjoy: https://t.co/vDSwVQYKbT

[Matt Pocock @mattpocockuk]

I was an AI coding skeptic until I tried plan mode.

Without it? Slop.
With it? Me-quality code.

We'll cover:

- What plan mode is
- How it helps the agent
- How it helps you
- Tips for improving plans
- "But surely coding solo is faster?"

[56kode @56_kode]

@UiSavior Be careful with right (contrast and color that matches the image)

[56kode @56_kode]

@UiSavior A of course

[Vercel @vercel]

We just released 𝚛𝚎𝚊𝚌𝚝-𝚋𝚎𝚜𝚝-𝚙𝚛𝚊𝚌𝚝𝚒𝚌𝚎𝚜, a repo for coding agents.

React performance rules and evals to catch regressions, like accidental waterfalls and growing client bundles.

How we collected them and how to install the skill ↓
https://t.co/kfLSbKl15X

[56kode @56_kode]

@vercel I love it, thanks

[56kode @56_kode]

Last week I asked #Gemini for a summary but forgot the URL. It gave me a summary anyway.

Apparently, prompt engineering is obsolete; we’re moving straight to telepathic engineering. Or just pure hallucinations. 😅🔮#AI

[56kode @56_kode]

Forget “Reasoning Avg” and “Coding Avg”. My only #AI benchmark: insults per day. #Gemini 3 Pro & Claude #Opus? So good I sometimes go 24h without calling them names. That’s basically AGI in my house.

[56kode @56_kode]

Our Vitest tests took 22s to start. Now? 4 seconds. 🚀

How Gemini 3 Pro identified "God Modules" killing our performance and automated the fix—saving us 15 min per CI run.

👉 https://t.co/dmdMmGIWtT

#Vitest #CI #react #reactjs

[56kode @56_kode]

🚀 Debug web apps directly from #Cursor #IDE with Browser-Tools #MCP! ⚡

🔹 Console logs analysis
🔹 Network requests inspection
🔹 AI-powered debugging

👉 https://t.co/Et3UEG4wum
#CursorIDE #AI #debug #chrome