An AI assistant for business is only as useful as your ability to answer it. Thirty-odd agent windows run on a Mac at home; I'm not at home. The hard part isn't the small screen — it's that two ways of connecting look identical and behave in opposite ways when you exit.
Most AI productivity tools are one more app to alt-tab to. These two are the opposite: a terminal file browser and a real Chrome that live inside the workspace your agents already run in. What broke on three machines, what we pinned, and the half-hour version.
Nothing about the tooling changed. What changed is that the conventions living in one person's head now have to live somewhere three people can reach.
Content production, video, publishing workflows
Your agent starts every session from zero because nothing about how you work is written down. An 8-layer knowledge base — brand, memory, workflows, tools, specs — plus two full production demos where only the knowledge base changes and the output changes completely.
The code licence and the weights licence are two documents, and usually only one of them gets read. Sort the field into synthesis, cloning, and conversion first, then let commercial terms cut the list before an ai assistant for business ever speaks.
Revisions eat the schedule because text-only generation hands back a middle nobody chose and ends that wander off the board. Ten keyframe prompt templates lock both endpoints up front, so what lands in between is a decision instead of a surprise.
An unused feature bills you in wasted renders, one hopeful attempt at a time. Constraining a request turns output from a lottery into a process, which is why the cheapest place to automate business processes is inside software you already pay for.
Tool-specific notes have the lifespan of the tool, and here that runs about eighteen months. A brief written one layer up survives the swap, porting across Runway, Kling, Veo and Seedance. Vendor independence is the unglamorous half of any attempt to automate business processes.
A catalogue shot nobody commissioned for video is still a shot you already paid for. Animating one costs a render; generating a product from nothing costs a wrong product on screen. That asymmetry makes reused assets the safest place to automate business processes.
When the crew stops arriving, so does the judgement that used to arrive with it. Nobody budgets for that, because the line item never changed. The three layers here are where craft gets written into the brief instead, the moment every field hits once it starts to automate business processes.
A fix you cannot start until you are back at a desk competes with everything else waiting there, and usually loses. Specifying a production system out loud, from a phone, removes that queue. It also quietly changes who inside a company can automate business processes at all.
Five Claude Code Skills, one unattended content pipeline: persona cloning, sourcing, drafting, image generation, and scheduled publishing. A workflow automation pattern built on single-responsibility design and file-based handoffs — each step reads exactly what the last one wrote.
Content teams rarely stall on writing, they stall on supply. This Claude Code Skill takes one topic, pulls real discussions, extracts non-obvious angles, and returns platform-ready drafts: 125 posts to 24 pieces in about ten minutes, with the full pipeline.
Battle-tested Ghost CMS practices from 18 months running two self-hosted sites: VPS deployment, Cloudflare DNS, theme work, newsletter growth, SEO configuration and multi-site ops, including the Admin API that lets ai automation tools publish without a human in the editor.
When I rebuild one with AI agents, you get the write-up — including the parts that didn't work. No weekly roundup, no "5 tools you need."