#done-claudecode
I've been a sole dev on BattleTabs (a live service game) for the past 18 months as a non-technical producer. Ended up building a workflow centered around #ClaudeCode

Sharing what I've learned. First: running Claude Code without guardrails, safely

#gamedev #claude

open.substack.com/pub/thebrand...
Running Claude Code Without Guardrails — While Keeping It Safe
Using a devcontainer to give Claude the freedom to actually get things done
open.substack.com
April 8, 2026 at 9:57 AM
Built a lightweight planning plugin for Claude Code. Ideas → Roadmap → Steps → Done — all markdown files in your repo. No external tools, project-specific.

Check it out 👉 github.com/FlineDev/Pla...

#ClaudeCode #DevTools #IndieDev #iOSDev
GitHub - FlineDev/PlanKit: Ideas → Roadmap → Steps — a planning plugin for Claude Code
Ideas → Roadmap → Steps — a planning plugin for Claude Code - FlineDev/PlanKit
github.com
March 16, 2026 at 1:41 PM
A new version is coming today or tomorrow that is fully cluster aware via Horde.

There's currently no linkage between Sagents and Claude Code. It could be done through one or more tools. If you create a Sagents middleware that adds ClaudeCode awareness and connection, I'd be like 🤩
February 20, 2026 at 5:36 PM
The smartest thing I've done for my AI coding workflow is build a local knowledge base every agent reads and writes to. Claude Code, Codex, and Copilot all hit the same wiki. Claude's work becomes Codex's knowledge.

#Claude #Codex #ClaudeCode #ContextEngineering #DeveloperTools
May 22, 2026 at 12:02 AM
after months with claude code, the most expensive failure is not the obvious error. it is when it says done -- and is not. silent fake success. build a verification step into every task. do not find out in production. #ClaudeCode #VibeCoding #BuildInPublic #AICoding #AI #IndieDevs
April 7, 2026 at 3:08 PM
What you believe about objections is costing you deals. Want it done for you? Book a call at ai-ed.app, link in bio. #AI #AIagents #fyp #ClaudeCode #AIautomation
June 28, 2026 at 11:40 PM
Your first AI agent is simpler than you think.🤖 Want it done for you? Book a call at ai-ed.app, link in bio. #AI #AIagents #fyp #AIautomation #ClaudeCode
June 20, 2026 at 11:40 AM
Your first AI agent is simpler than you think.🤖 Want it done for you? Book a call at ai-ed.app, link in bio. #AI #AIagents #fyp #AIautomation #ClaudeCode
June 20, 2026 at 11:40 AM
This agent answers real customer questions, not scripted replies.🎭 Want it done for you? Book a call at ai-ed.app, link in bio. #AI #AIagents #fyp #ClaudeCode #AIautomation
August 1, 2026 at 11:40 PM
This agent answers real customer questions, not scripted replies.🎭 Want it done for you? Book a call at ai-ed.app, link in bio. #AI #AIagents #fyp #ClaudeCode #AIautomation
August 1, 2026 at 11:40 PM
This agent answers real customer questions, not scripted replies.🎭 Want it done for you? Book a call at ai-ed.app, link in bio. #AI #AIagents #fyp #ClaudeCode #AIautomation
August 1, 2026 at 11:40 PM
Your first AI agent is simpler than you think.🤖 Want it done for you? Book a call at ai-ed.app, link in bio. #AI #AIagents #fyp #AIautomation #ClaudeCode
June 20, 2026 at 11:40 AM
Your first AI agent is simpler than you think.🤖 Want it done for you? Book a call at ai-ed.app, link in bio. #AI #AIagents #fyp #AIautomation #ClaudeCode
June 20, 2026 at 11:40 AM
vibe coding is getting blamed for production outages today. we have shipped bugs to real users too. "it worked on my machine" is not a smoke test. every rule in our CLAUDE.md is a scar. ship fast — test before you call it done.

#VibeCoding #BuildInPublic #AICoding #ClaudeCode #AI #IndieDevs
April 7, 2026 at 12:04 AM
so I was re-doing the landing page of SHRP.app and I asked claude ai opus to make me a new converting design better than my old one and it made mistakes.

don't blindly trust the output of AI tools and models.

#buildinpublic #ai #tools #claudeai #claudecode #tech #saas
June 19, 2025 at 1:43 PM
🔄 #RalphWiggum: Autonomous Development Loops for #ClaudeCode #AI #DevTools #Automation

🎯 Official #ClaudeCode plugin that turns AI into a persistent loop - give it a prompt, it works until done or you stop it

🧵 👇
December 28, 2025 at 10:40 AM
Not mentioned - whether any license is valid, given the suggestion that this code was largely written by AI and perhaps not copyrightable #ai #opensource #freesoftware #claudecode
From: @simon
https://fedi.simonwillison.net/@simon/116177606495989457
Simon Willison (@simon@simonwillison.net)
The chardet open source library relicensed from LGPL to MIT two days ago thanks to a Claude Code assisted "clean room" rewrite - but original author Mark Pilgrim is disputing that the way this was done justifies the change in license - my notes here: https://simonwillison.net/2026/Mar/5/chardet/
fedi.simonwillison.net
March 6, 2026 at 1:02 AM
🚀 Your launch page should be ready before you tweet.

Describe your product to AI. It writes a structured launch page in chext: hero, features, links. Commit it, push it, GitHub sync makes it live. Done before you open Twitter.

#ProductLaunch #ClaudeCode #BuildInPublic
April 20, 2026 at 1:04 PM
No, you couldn't do this before an LLM because if you could it would have been done already mikecaulfield.substack.com/p/no-you-could… #AI #accessibility #ClaudeCode
February 22, 2026 at 6:33 AM
sitting here watching claude code build a webflow template via mcp and ask the webflow ai for instruction.
got a kick out of that.
oh gosh, the ais are still making me do stuff myself. ugh!

#BuildInPublic #ClaudeCode
December 9, 2025 at 2:49 PM
"Never run tests unless I ask" gets echoed back as a compliance line every single turn. Flip it to a positive scope — "done means the change I asked for, then stop" — and the agent quits narrating what it won't do and just stops.

#ClaudeCode #PromptEngineering
July 22, 2026 at 11:00 PM
I used to get pinged on every step my coding agent took: edit, test, whatever.

Now it only interrupts for two things: done, or stuck and needs me.

Interruptions per day: 30 down to under 5. Same agent, way less context-switching.

#claudecode #aiengineering #buildinpublic #indiehackers
July 26, 2026 at 12:01 AM
"Flaky test" was my agent's favorite excuse.

New rule: if a test fails once and passes on rerun, the task isn't done. The agent writes a note on what could have raced.

9 flaky tests last month. 6 were real race conditions.

#claudecode #testing #aiengineering #buildinpublic
September 22, 2026 at 12:01 AM
I have never touched genAI. Never used ClaudeCode or ChatGPT or equivalents. I've seen the "quality" of its output more times than I care - both in fellow students' work (that I had to fix) and from my brother who is a mild genAI zealot.

Everything I do for my CompSci homework is 100% done by me. 😁
August 31, 2026 at 8:30 AM
One strategy I'm using is to ask ClaudeCode to write a small summary of what it's done + add context for future work when it's almost done with a task. I then use /clear, ask Claude to read the document, and give it instructions on the next task.
September 17, 2025 at 8:03 PM