Automation Studio
Describe it, pick a flow, watch it ship — with your models, on your machines or ours.
The Automation Studio
0:38
An idea described in plain words, and the flowchart actually running it.
What the studio is
The Automation Studio (localkeep.ai/automation) turns a described idea into a watched pipeline. You write — or dictate — what you want, pick a flow, choose an AI model for every step, choose where each step runs, and press launch. The flowchart then lights up step by step: brainstorm, plan, develop, test, review, deploy. You can sit back and watch, or step in at any moment.
The flows
•
Website — idea to deployed site, ending with a domain-and-launch runbook.
•
Backend / API service — idea to running service with an operations runbook.
•
Mobile app — idea to store-ready build with a store-submission runbook.
•
Document / report — outline, full draft, edit and polish.
•
Video / script — concept, full script, shot list, production runbook.
Every flow ends at production reality. Steps marked RUNBOOK cover the parts of shipping that live in other companies' consoles — domain registrars, App Store review, DNS. For those, the studio generates a precise, personalised checklist with exact URLs, exact copy to paste, expected costs and durations. It will not pretend to click Apple's buttons for you; it will make doing so yourself a ten-minute follow-along.
Models and machines, per step
•
Each step can use a different model — a big model for planning, a fast one for boilerplate.
•
Cloud steps run on Local Keep AI and meter on your plan or API credit.
•
Device steps run on YOUR machines: install the CLI (pip install local-keep-ai-cli), log in, and the machine appears as a target. Its local models do the work at no metered cost.
•
If a different model answers than the one you picked (capacity), the step says so — substitution is always disclosed.
Editing a live run
•
Queued steps: change the model, the target, or add instructions any time before they run.
•
The active step: Stop it — the run pauses, nothing is lost, and your edits apply when it re-runs.
•
Done steps are history: their output stays viewable and feeds the steps after them.
•
Resume continues exactly where the run stood.
What it costs
Cloud inference meters like any other usage on your plan or API credit. Hosted cloud resources provisioned for a run are billed at metered cost with a disclosed markup. Steps on your own devices cost nothing beyond your electricity.
Last updated: 25 August 2026
All documentation
Installation
Install the CLI on macOS, Linux or Windows, and sign in.
Setup, start to finish
The whole path on one page: install, sign in, pick local or hosted, get an answer.
Connecting accounts
One account everywhere, registering devices for the SMS bridge, and OAuth integrations.
Models
Pull open-weight models, switch between them, and run fully local.
Using models on every client
What runs locally in a terminal, a browser tab, a phone and the desktop app.
Commands
Every terminal command and every slash command, each with its own page.
Editor integrations
VS Code, Cursor, JetBrains, Neovim, Sublime, Zed, Godot, Unity, gh and git hooks.
The SMS bridge
Send a task to your own machine from your phone, and get the answer back.
The apps
One React Native codebase behind iPhone, Android, web and desktop — and how to run it.
Enterprise seats
Licensed team seats under one owner account — invites, billing, and how seats resolve.
API reference
API keys, authentication, chat completions, pricing and errors — for your own software.
Notifications
Four channels, all opt-in, how to change them, and the limits on re-engagement.
Privacy
What leaves your machine, what never does, and how anonymisation works.
Troubleshooting
When a model will not load, a command hangs, or the CLI cannot be found.