Introducing General Input Specialists

General Input now runs on AI specialists, one for each job in your business. Here's why we stopped asking you to build automations and let you delegate.

Introducing General Input Specialists

General Input now runs on AI specialists. Each specialist owns one job in your business, such as weekly reporting, lead follow-up, or customer win-back. You tell it what you want done in your own words, and it builds and runs the skills, workflows, and apps behind that job. When the job changes, you tell the specialist, the same way you would tell the person who owns it.

If you already use General Input, everything you built is still here. The difference is that every piece of it now has an owner you can talk to.

We were asking you to build software

People come to General Input to get work off their plate. Until now, the first thing we asked in return was a set of builder's questions. Should this be a workflow or an app? What triggers it? What should it be called, and where should it live?

Anyone who has handed a job to a capable colleague knows delegation doesn't work like that. You describe the result you want, answer a few questions, and let them figure out the rest.

Specialists work that way. Tell one, "Every Monday, send me revenue by plan and flag anything that dropped more than 10%." It asks where your revenue data lives and who should get the report. Then it builds what the job needs: a skill that records how the report is put together, a workflow that runs it every Monday morning, and an app where your team reads the results. You describe the job once, and the specialist works out which pieces it takes.

Automation needed an owner

The systems that worked best on General Input were never a single workflow or app. They were a few of each, working toward one goal, and they stayed useful because someone looked after them. That person noticed when a report came out wrong, remembered how the pieces fit together, and updated them when the process changed. Usually it was you.

A month ago we launched projects to give those systems a home. Projects kept things organized, but a folder can't notice a failed run or ask what you meant. They also left an awkward question: who do you talk to? You could ask a general assistant, the assistant on each workflow and app, or a chat inside the project, and each one saw only part of the picture.

A specialist answers both. It owns the job, so there is one place to ask for a change, and it's the one that comes to you when something needs a decision. Its conversations, memory, files, and work all live on one page.

Why AI specialists build systems you can see

There's a version of AI delegation where an agent does the job from scratch every time, out of sight. It demos well, and it's hard to trust with anything that matters, because nobody can check how the work was done.

A specialist turns what you teach it into things your team can inspect. Skills write down how the job is done. Workflows run on schedules and triggers the same way every time, and every run is recorded. Apps put the results where everyone can see them. And the specialist's memory, which holds its rules and the decisions behind them, is plain language that everyone you share it with can read.

That's also what separates a specialist from a chatbot. A chatbot works while you prompt it and forgets the job when the conversation ends. A specialist's work keeps running between conversations, and what it learned about your process stays written down, so the process no longer lives in one person's head.

Why one job each

We call them specialists because each one is meant to do one job well. We've written about why a dozen small agents beat one giant one: small ones are easier to understand, to hand to a colleague, and to fix when the work changes. So when a new job comes up, it gets its own specialist.

Every member starts with one, their Chief of Staff. It holds anything that doesn't have an owner yet and can set up a new specialist when a job deserves one. If you'd rather start from a working example, browse specialist templates, each built around a single job.

What changes if you already use General Input

Nothing you built was deleted or turned off. Each project is now a specialist, with the same workflows, apps, skills, and files, and everything outside a project is with your Chief of Staff.

The next time you sign in, a short tour walks you through the change. If you have work to sort, it offers to split it into specialists, one per job, and shows you the proposed split before anything moves.

If you have questions or want help setting up your first specialist, book a call with Jack.

FAQs

What is an AI specialist?
An AI specialist is an AI agent that owns one job in your business, such as weekly reporting or lead follow-up. In General Input, it builds and runs the skills, workflows, and apps that job needs, keeps a written memory of how the work is done, and is who you talk to when the work should change.
How is an AI specialist different from a chatbot?
A chatbot works only while you prompt it and forgets the job when the conversation ends. An AI specialist keeps its work running on schedules and triggers, writes down how its job is done, and comes to you when something needs a decision.
How is an AI specialist different from workflow automation?
With workflow automation, you design every step and trigger yourself. With a specialist, you describe the outcome, it asks what it needs to know, and it builds and maintains the workflows, apps, and skills behind it.
What happened to projects in General Input?
Every project became a specialist, with the same workflows, apps, skills, and files, so nothing needs rebuilding. Work that wasn't in a project is with your Chief of Staff, and a short tour offers to sort it into specialists, showing you the split before anything moves.
How much work should one AI specialist handle?
One job, the kind of work one person would own at a desk: a channel, a workstream, a recurring report. When a new job comes up, give it a new specialist.
Can my team see my conversations with a specialist?
No. Your conversations are private to you, even when the specialist is shared. What it knows lives in its memory and files, which everyone you share it with can read.

Hand your first job to a specialist.

Describe the outcome in your own words. Your specialist asks what it needs to know, then builds the work and keeps it running.

Create your first specialist

Free account. No credit card required.

Explore more posts