Three days into Implementation Playbooks. I want to know what came out of it. Not what you watched. Not what you understood. What did you actually MAKE? - An animation you finally got working? - A folder structure that finally makes sense for your AI setup? - A browser workflow that's saving time? - Or something else entirely? Drop it below. Doesn't matter how small. Some of the most useful things in here started as the smallest outputs. And if you got stuck on a step... Don't sit there spinning. - Bring it to the Clief Notes AI - Bring it to the next call - Go back to the lesson and figure out WHERE you got stuck. That's part of the build. The roadmaps aren't here to make you feel like you completed something. They're here to make sure you walk out the other side with something you can actually use. Drop what you built below. 👇

I'm curious: who here is using Obsidian for their knowledge management and integrating it with an ICM setup? I'm looking to combine my local markdown vault with AI agents to build a permanent context layer, and I'd love to get your thoughts: 1. How do you integrate them? (e.g., using the MCP filesystem server, specific Obsidian plugins, or custom local scripts?) 2. What is your honest opinion on this? Is it seamless, or does the AI get bogged down once your vault grows and files get messy? Any tips on folder structures (like separating an /Inbox from a /Core folder) or system prompts that keep the agent from messing up your notes would be highly appreciated. Thanks!

Ran a four-stage audit pipeline (my first real ICM build), assuming each stage got a fresh window. Nope. After the full run: 302.8k tokens — 281.3k of it transcript. Everything I engineered (CLAUDE.md, contracts, skills) was 25k. My architecture: 8% of the window. The chatter: 92%. One /clear: down to -282k. Messages: 137 tokens. It picked the run back up from the stage output folders without asking me anything, because my CLAUDE.md tells a fresh session to orient from artifacts, never from the conversation. (Protocol block in the comments if you want it.) So the files hold the state, but to run /clear, I have to do it between the stages? How are y'all automating the clear between stages? By hand at every gate? A wrapper running each stage as its own `claude -p`? Something else? Or does it not matter?

Say you've got a plan review, a brand doc, a spec - whatever - and two or three other files were built off it. Now the source updates. Nothing tells the downstrem files they're working off old data. No error, no flag, just quietly wrong output until someone catches it by accident. How is everyone actually handling this in their systems: - pin every dependency to a version and force manual re-approval when the source changes? - some kind of "last verified against version X" field sitting in front matter? - or honestly just not solving it and eating the occasional bad output? Genuinely curious what's held up for builders running this past a handful of files...

I’ve just seen Thariq on X post this and well, yes, thank you, we are already doing this https://claude.com/blog/the-new-rules-of-context-engineering-for-claude-5-generation-models