AI assistants and search decision
Jared Stivala
A competent developer can build a usable macOS AI-autocomplete replacement (prior open-source projects exist), but reproducing the polished, signed distribution, support, and bundled model access of a paid product is non-trivial.
Visit website↗Not priced
No pricing page we fetched carried a figure, so there is nothing to compare against. The build side is still real.
$100one-off90 h to build
$50/mo3 h/mo upkeep
No published price to break even against.
Open-source builds that already do this
Every project below is open source and already does this job today. Fork one, self-host it, or take the parts you need - the build prompt further down assumes an empty file, and this is the shortcut past that. Licences differ; check the one on each card before you ship. All Jared Stivala alternatives, with the arithmetic →
What a replacement has to do
- Intercept typed text system-wide, send context to an LLM, receive completions, insert completions inline, allow user to accept/edit.
What it still won’t have
- Polish and UX refinements (edge-case input handling across mac apps)
- Officially signed/notarized app distribution and update infrastructure
- Customer support, analytics, and usage-driven model tuning
- Any proprietary server-side optimizations, telemetry, or bundled model access
What remains hard
- Product polish and ongoing maintenance
First-year cost
No published price
Jared Stivala does not publish a price we could read, so there is nothing to compare against. What building costs is below.
Money you would actually spend
Time you would spend
—
What you would spend
What we assumed
The verdict above measures whether you could build it. This one is only about money.
Runnable build prompt
Build a macOS menubar AI-autocomplete app using Swift + SwiftUI for the UI and a small local SQLite store. Integrate with an external LLM via HTTPS (e.g., OpenAI-compatible API). Core features in scope: system-wide keyboard listener using Accessibility APIs, context capture (last N tokens per window), prompt construction, async API calls with timeout and caching, inline completion insertion with undo, basic settings UI (toggle, model API key, max tokens), and auto-update check. Out of scope: proprietary model training, multi-device sync, paid billing system, advanced telemetry. Include error handling for API/network failures, accessibility permission checks, unit tests for prompt construction and insertion logic, and CI build scripts for notarized packaging.
How we checked
How the score was reached
- Partly verdict base52
- An open-source build was found+5
- 3 cited sources+3
- Evidence score60
The base comes from the verdict. Everything under it is a check that either happened or did not, and each one is a fact frozen in this record rather than a judgement made at render time - so the same evidence always produces the same number.
How scoring works →Cited sources · 3
Every page the run actually retrieved.
- official productFriendware - AI Autocomplete for macOS
- open sourcejohnbean393/KeyType
- open sourceFuJacob/cotabby
Integrity checks
What held up, and what did not.
