I see this question a lot, so I wanted to share my take. ICM (the stage-by-stage folder setup, where each stage loads only the context it needs) cuts waste. It doesn't cap usage on its own. If your usage is still high, it's usually one of these: **1. Your context files are too big.** This is a symptom of ai generated slop without proper human auditing, maintain your files! If each stage loads a 5,000-word reference doc "just in case," you've moved the bloat into a folder. Check what each stage actually reads. Keep stage instructions short and push reference material into files that are only loaded when needed. **2. Stages aren't isolated.** ICM only saves tokens if each stage starts clean. If you run everything in one long conversation, the model re-reads the whole history on every turn. Start a fresh session per stage and pass work forward through output files, not chat history. **3. Too much gets loaded at the start.** A bloated root `CLAUDE.md`, or a pile of always-on rules, skills and MCP tools, is paid for on every single message. Trim the always-on layer to the essentials. Manage your hooks and permissions properly! **4. Outputs are being re-read in full.** If stage 3 reads the entire raw output of stages 1 and 2, you're paying for it again. Have each stage write a short summary or handoff file, and have the next stage read that instead. My /handoff skill is one of my favorite and most used. One of the few things that's persisted in all my "restarts". **5. Your model choice is off.** Running the strongest model on every stage burns usage fast. Use a lighter model for mechanical stages (formatting, extraction, file moves) and keep the heavy model for the stages that need judgment. !! I was just burned on this today. I had my fancy new process all shiny and polished. Yesterday I was polishing and left a mark - today, Opus and Fable CRUSHED my account usage before I could notice, and I feel like I'm a careful person. **6. Loops and retries.**

Hi Guys id be really interested to see if anyone is using GROK as the orchestrator and how theyre directing it using the ICM method that Jakes teaching us about, im a novice user and just started my journey, Im trying to direct it to use my claude code subscription for the lessor or other agents, and reduce token usage, but Kelly (my chief of staff) which uses grokbot harness is my orchestrator, , any videos would be greatly appreciated or help. thanks in advance.

I've noticed something lately — there's this weird barrier when I'm reading AI-generated text. I just don't want to read it. But if it's human-written? No problem. It's not that the AI is giving me bad information. It's more like... the writing style itself creates friction. Too polished, too formal, or just predictable in a way my brain picks up on. I'm curious if anyone else is running into this, especially if you're using AI for client work, internal docs, or anything people actually have to read. And if you've found ways to fix it — whether that's prompt tricks, editing workflows, or just accepting that AI is better for structure than final copy — I'd love to hear it.

I have been using Claude Code, Cowrk, and Chat for more than 2 years. I built my existing folder structure before Cowork was even a thing, so all of my AI project folders are on my root folder where Documents, Desktop, Downloads, and the other system folders are located. By the time Cowork was released I had already built several apps with repos, databases, and deployments that relied on my current folders staying right they were or I would have to rebuild everything. Now I'm trying to modify this mess of a structure, which as a global ~.claude/CLAUDE.md file that is now over 1MB, which by markdown file standards is the bloat of all bloats. I just updated the icm-architect skill to use AGENTS.md files instead of CLAUDE.md, but in the global hidden folder, this does not work, because it is ONLY Claude specific. So, as I'm making these changes I'm getting pushback from Claude. The biggest is routing. Here's my question. Claude code uses the .claude folder, so I can't get around this. Should the CLAUDE.md file be a duplicate of the AGENTS.md at the ~root folder, in my case? Right now my ~/AGENTS.md file has 11 global rules, mapping to all the project folders, and instructions to use the icm-architect skill for creating new projects.

Serious question for everyone running Claude, ChatGPT, AI agents, or automated workflows: **What if I could help you reduce your token consumption by 30–70% without sacrificing output quality or productivity?** Not by simply using fewer features, but by analyzing and optimizing your actual system architecture, agent orchestration, context management, model routing, and workflows. I'm developing something called **THE GERMAN STANDARD — German Agentic Systems Assurance (GASA)**. A structured technical audit and optimization framework based on established engineering principles and aligned with relevant international ISO standards. The idea is simple: 1. Audit your existing AI setup. 2. Identify inefficiencies, bottlenecks, and unnecessary token consumption. 3. Establish a measurable performance baseline. 4. Optimize your system. 5. Run identical benchmark tasks before and after. 6. Document the actual savings and verify that quality hans't declined. The 30–70% range is a proposed optimization target, not a guaranteed result. Actual savings would depend on your setup and would need to be demonstrated. **Now here's my question:** If I could demonstrate a measurable 30–70% reduction in token consumption on your actual workloads, with comparable results, what would that be worth to you? Would you prefer a one-time audit fee, a complete optimization package, or a performance-based model where I receive a percentage of the money you actually save? I'm not asking what you'd pay for another AI course. **I'm asking what you'd pay for a measurable improvement to a system you're already spending money on.** Would love to hear actual numbers.