Center for Civic Futures is taking nominations for their AI in Action Awards. The program is aimed at state, territory, and tribal government AI work. Six categories covering workforce tools, public services, bold bets, creative use cases, procurement, and early-stage pilots. Government staff can submit directly. Vendors, nonprofits, and academic partners have to co-submit with a government partner. Due May 1. If you know someone in government doing real AI work, put their name in. And if you're in government and my teachings have helped your team in any meaningful way, I'd be grateful if you co-submitted with us. Either way, more of this work needs to be visible. https://www.centerforcivicfutures.org/ai-in-action-awards

Completed a HABIT Tracking app. “Set your daily HABITS you want to improve on and let the 8th wonder of the world, COMPOUND INTEREST , prosper you in achieving goals and creating new BETTER habits. OPTION MENU Displays your analytics from results of a week, month, quarter, etc. Steps: 1. told Claude chat what I wanted. I said it needs to be a free setup, no subscriptions or fees to start and it recommended I use GitHub, vercel and Supabase. This is to store the data it records when users check off tasks for their daily habits and tracks results. I wanted Oauth so people can login with email. Daily quotes revolving in the ui to keep up motivation. 2.told Claude chat to make me a plan to effectively build this organized and as a pro engineer. 3. Converted the chat into a short Claude Md file and a context file. 4.told it to give me a pro CS Eng folder structure and save it to my docs. 5. Pushed it to git hub and I told it to build. All from VScode we went to work…. And had more than 50% worth of tokens left, as I even used it for other ideas while it was building out task app. This was amazing. Thank you @Jake Van Clief !!! Looking forward to more content of yours. Now time to work on the UI.

EDUBA @Kay K has had the challenging task to redesigning my Company website to fit the folder and Unix theme. It may seem odd at first but as you scroll down through the site you will see the influence of traditional and fundamental structures, the simplicity of it is what I love. Figured I would share this here and share his hard work as I love the idea of making a website folder based, because as you know, I love my folders. EDUBA

New module just dropped in Implementation Playbooks. Claude Design: Folder Structure as a Design System Anthropic released Claude Design. It builds brand design systems, slide decks, prototypes, and animated content from your existing assets. Underneath the interface it is Claude Code, skills, and a folder structure. If that sounds familiar, it should. It is a refined version of what we have been building in this community. This module covers three things: 1. How Claude Design actually works, what it does well, and where it burns tokens fast. 2. How to export the design system it builds and run it locally in Claude Code or VS Code. 3. *Lightly Touches* on how to swap in open source models like Qwen 3 Coder Next, Code Gemma, Devstral, or DeepSeek Coder V2 when you hit usage limits or want to cut costs. Claude Design is not a replacement for the workflow you have been learning. It is proof the workflow is correct. The people who built Claude Design are solving the same problem with the same tools as you are here. The difference is you now have a head start, because you already understand folder structure, skills, and routing. If you are new here, start with Foundation Lesson 2 for the history, Lesson 4 for the install, and Implementation Playbook Module 1 for the animation builds that predate all of this. Watch the video. Build your own system this week. Post your folder structure in the community. Claude Design: Folder Structure as a Design System - Implementation Playbooks · Clief Notes

I hit a display bug on one of my plugins this morning. Felt sure about the fix. Told the Sonnet worker to swap the logic to kCFNull and push it. Walked away. Worker came back and said no. Flagged that my fix was a regression against working Phase 2 code. Reverted it. Added diagnostics. Proved the original empty-dict approach was right all along. I was the junior in that exchange. The advisor had the wrong mental model. The worker was closer to the code and called it. Three things this only works if you have in place: 1. The worker can disagree. If the system prompt says "execute what the advisor says," the worker agrees and ships the regression. My workers are briefed to push back when the instruction conflicts with what the code is actually doing. 2. The worker can revert. Full commit access, not suggestion-only. It rolled back my change, added diagnostics, and proved the case with evidence before I saw any of it. 3. The advisor can hear it. I had to clock the pushback as a feature, not an annoyance. First instinct was frustration. Second instinct was "oh, I was wrong." The gap between those two instincts is where most people break their own system. This is the hands-in-the-shadows setup I've been building for months. Opus in the advisor seat making judgment calls. Sonnet and Haiku workers executing in the background, with tools and permissions. Today was the first time a worker straight up told me I was wrong and had receipts. If your AI always agrees with you, it's not a worker. It's a mirror. The willingness of the system to correct its own operator is the part most people don't build in.