We make four families of small apps. They all run on your own computer and open in your browser, and they all need free Node.js 18 or newer on Windows, Mac or Linux. What differs is the job, the AI service behind it, what a run costs you, and where your text goes. Please read the row for your family before buying.
Thirty-second version
A letter, a policy page, a study guide, a meeting action plan, downloaded as a Word file, with "Why I suggested this" and "Check these yourself" notes.
Kitson Pro · $29 eachSpreadsheets, data sheets, manuscripts, reference lists, requirements and test results. For engineers and scientists. Early release.
App Maxxing · $39 eachCustomer emails, leads, reviews, expenses, tasks or bug reports, sorted in seconds for about a thousandth of a cent per check at launch prices.
Quick sorting apps · $13 eachPrivate mode uses free Ollama on your own machine (8 GB+ RAM). Or use your own AI provider key in Cloud mode.
Private AI apps · $16 to $21Then none of these apps is for you yet, and that's fine. Start with our free guides and printables instead.
Free downloadsFrom each listing, October 2026
| Kitson Pro | App Maxxing | Quick sorting apps | Private AI apps | |
|---|---|---|---|---|
| Apps | 10 | 10 (early release) | 10 | 7 (early release) |
| Price | $29 each; all 10 $249 | $39 each; all 10 $335 | $13 each; all 10 $110 | $16 to $21 each |
| AI service | Claude Opus 5.5 (switchable to Sonnet 5.5) | OpenAI models | Jev, a fast decision model from TypeSafe AI | Free Ollama on your computer, or your own AI provider |
| What you need | Your own Anthropic account and key. A Claude chat subscription does not include it. | Your own OpenAI account with API access and a key. A ChatGPT subscription does not include one. | Your own API key from TypeSafe (may have a waitlist) or Vercel AI Gateway | Ollama with 8 GB+ RAM (Private mode), or your own AI provider key (Cloud mode). No AI model is included. |
| Running cost (our tests) | About 4 to 30 cents a run, paid to Anthropic | About 3 to 35 cents a run on short sample files, paid to OpenAI; longer files cost more | About a thousandth of a cent per check at launch prices, paid to your provider | Nothing extra in Private mode; your provider's prices in Cloud mode |
| Where your text goes | To Anthropic, under its terms. Not to us. | To OpenAI, under its terms. Not to us. | To TypeSafe (or through Vercel), after the app removes email addresses, phone numbers and card-like numbers | Stays on your computer in Private mode; to your provider in Cloud mode |
| Cost controls | Cost shown before and after each run; you set a limit for one run | Cost shown before and after each run; you set a limit for one run | — | — |
| Watch first | 10 free videos | 10 free reviews | 10 free videos | See the app list |
| If you can't get it running | Email within 14 days: help, or a refund of the app price | Email within 14 days: a refund of the app price | Email within 14 days: help, then a full refund if we still can't get it working | Email within 14 days: help, or a refund |
Running costs are our own measurements on sample inputs and depend on your documents and on each provider's prices, which can change. Kitson Workshop is not affiliated with or endorsed by Anthropic, OpenAI, TypeSafe AI or Vercel.
$29 each
Contract Explainer, Job Application Tailor, Business Plan Reviewer, Complaint Letter Writer, Study Guide Maker, Spreadsheet Formula Builder, Grant Application Helper, Small Business Policy Writer, Meeting Action Plan and Document Q&A. Guided questions and a built-in sample, so you never face a blank box. When an app quotes your document, it checks the wording and marks any quote it cannot find.
Not for you if you need professional advice: these apps explain, summarise and draft, and what they write can be wrong.
Bought separately the ten cost $290. Code SAVE10 takes a further 10% off the package ($224.10).
Early release · $39 each
Spreadsheet Formula Checker, Data Sheet Comparer, Statistics Reporting Checker, Revision Change Register, Test Results Against Limits, Reference List Checker, Long Report Consistency Checker, Evidence Table Builder, Requirements Trace Matrix and Reviewer Response Matrix. The apps point things out for you to check: they do not certify, approve or sign off anything.
Please read: each app has passed our own offline tests and has been run on its built-in sample against the live service (October 2026), but has not yet been tried on a wide range of real files. You may meet faults. Fixes are sent to buyers free.
Bought separately the ten cost $390. Code SAVE10 takes a further 10% off the package ($301.50).
$13 each
Inbox Triage, Lead & Form Sorter, Review & Comment Moderator, Scam Warning Signs, Job Match Checker, Expense Sorter, Task Prioritizer, Email Tone Check, Bug Report Triage and Feedback Theme Sorter. Paste a pile of messages, reviews, tasks or transactions and get them sorted, with unsure rows marked for you to check, and a spreadsheet (CSV) download.
Not for legal, medical or emergency messages. Automated sorting can be wrong.
Code SAVE10 takes the package to $99.
Early release · $16 to $21
Code Explainer ($21), Reply Desk ($19), Meeting Notes ($19), Listing Writer ($19), Study Buddy ($16, ages 13+), Email Writer ($16) and Game Master ($16, ages 13+). Use free Ollama for Private mode (8 GB+ RAM), or your own AI provider key for Cloud mode.
These are early, educational releases: we're learning and improving them with every update, and AI answers can be wrong.
For educational purposes only: not legal, financial, tax, medical or other professional advice.