Claude for Marketing: Field Notes from an Operator

Setup: Chat, Cowork, Code and models

Claude just merged Chat and Cowork — the decision that disappeared, and the caveat that didn't Chat and Cowork are now one conversation with a full agent runtime underneath. What changes for operators: no more routing, deliverables inside the chat — and the local-tasks caveat most coverage skips. Updated 21st October 2026
Cowork reached my phone this week. I still can't tell where chat ends and Cowork begins. Anthropic put Claude Cowork on mobile this week and OpenAI shipped ChatGPT Work days later. The Code-vs-Cowork line is clean; the Chat-vs-Cowork line still isn't. An operator's honest map of a messy, powerful moment. Updated 11th September 2026
Claude Cowork keeps hanging? Here's the fix that runs every session. When Claude Cowork sits on "Working on it" or stalls in a decision loop, the message on screen tells you which problem you have. How to stop a stuck task, restart a session that won't start, and set the instruction that prevents most hangs. Updated 5th November 2026
Ultracode isn't a bigger thinking dial. It's Claude deciding how to split the work. I assumed ultracode was just /effort high with the dial turned up. It isn't — one changes how hard Claude thinks, the other changes what it does. Here's the difference, and why it's quietly carrying my parallel workstreams. Updated 21st October 2026
Fable 5 had us dazzled. But Opus 4.8 is quietly doing my work. Opus 4.8 needs a few more loops and a little more patience than the flashier model — but with the right context, the output quality is genuinely excellent. Why my Brain system is the unfair advantage. Updated 11th September 2026
Fable 5 is back. My terminal still hands me the junior. Fable 5 returned on 1 July — but Claude Code still defaults to Opus 4.8. A senior partner in one window, a junior developer in the other. Know which one you're talking to. Updated 11th September 2026
A context window is working memory, not a hard drive. Feel that, and you manage AI differently. The AI doesn't remember everything you've told it. It has working memory, and it's finite — here's the cheap trick I use to hear it filling up, and the fix when it does. Updated 11th September 2026

Credentials and safety

Skills, agents and scheduled routines

Ads and measurement

SEO and building small tools

The market around Claude

Microsoft's new Copilot looks just like Claude. That's not the part that matters. Microsoft relaunched Copilot with Home, Code and Autopilot. It looks a lot like Claude, but the real story is distribution: Microsoft is already inside the business. Updated 21st October 2026

Quick answers

What happened to Claude Cowork?

It didn’t disappear. Anthropic merged Chat and Cowork into one Claude: one conversation with the full agent runtime underneath, so you no longer pick a mode first. The catch: the runtime lives in the cloud, your computer doesn’t. Cloud tasks keep running when you close your laptop; tasks that touch local files still need the desktop app open. Read the full note.

Is ultracode just effort high turned up?

No. The effort setting changes how hard Claude thinks about one task. Ultracode changes what Claude does: it decides how to split the work. That is what quietly carries my parallel workstreams. Read the full note.

Claude is stuck on “Working on it”. What do I do?

The message on screen tells you which hang you have. Usually it isn’t thinking but stuck in a planning loop. Interrupt it (“Stop. Proceed with your best judgment.”), retry with a narrower first step, and set a short standing instruction so it makes a reasonable call instead of stopping to ask. Read the full note.

What is the difference between a Claude skill and an agent?

A skill is a recipe Claude pulls off the shelf when a task matches it. An agent is a coworker you hand a whole job to. The real difference is memory, not power. Read the full note.

How do I let Claude use my tools without giving it my passwords?

Give Claude access to the tool, not the secret. I put API calls behind small Cloudflare Workers that hold the keys, and use a password vault on my computer as the fallback. Read the full note.

How I keep this current

Claude changes every few weeks, and posts about it go stale fast. Once a month I check the most-read notes against Anthropic's own documentation, fix what changed and mark the post as updated. If something here is out of date, tell me.

Anthropic's documentation: Cowork and chat are one Claude · Scheduled tasks · Skills · Skills in Claude Code

Claude for operators