Frequently asked questions
Can’t find your answer? Email founder@triagekit.space.
The product
What is TriageKit?
TriageKit is a native macOS menu-bar app that watches the iOS Simulator while you build. It uses AI vision, through the model you choose, to spot UI bugs, traces each one to the source file responsible, and prepares a patch for you to review.
What kinds of bugs does it catch?
Visual ones: clipped or overflowing text, content hidden behind the keyboard, safe-area violations, overlapping elements, and rendering glitches. It does not replace unit tests, and it can’t catch logic bugs that have no visible symptom.
Which frameworks does it support?
Swift and React Native projects running in the iOS Simulator. After a patch is applied, TriageKit triggers a Metro reload for React Native or an Xcode rebuild for Swift, then checks the screen again.
Does it work with Android or real devices?
Not at the moment. TriageKit works with the iOS Simulator.
Can it change my code without asking?
No. Fixes are prepared as a git diff and shown in the menu bar. Nothing is applied until you approve it, and you can reject any patch.
How is this different from a terminal coding agent?
Terminal agents work from text: files, logs and command output. TriageKit adds the visual layer by capturing what your app actually renders, then connects what it sees to the code. It complements your editor and agent rather than replacing them.
AI models and tools
Which AI models can I use?
TriageKit works with Claude and all other major AI providers through your own API key. Pick the provider that fits your budget and your team’s data policies.
Can I run a local AI model?
Yes. You can run an AI model on your own computer and connect it to TriageKit. With a local model, TriageKit doesn’t send your screenshots or code to a third-party AI provider.
Does it work with Claude, Codex and Gemini coding tools?
TriageKit doesn’t plug into AI coding tools directly, because those tools don’t offer an integration for it. Instead it offers report handover: TriageKit prepares a report of what it found, and you hand it to Claude, Codex, Gemini or another tool to work on the fix.
Privacy and data
Is my code uploaded to TriageKit?
No. Project files and compiler logs are read on your Mac, and TriageKit does not run a server that receives your source code. When TriageKit analyzes a screen or proposes a fix, the relevant screenshots, accessibility data and file excerpts are sent directly from your Mac to the AI provider you choose. If you use a local model, nothing leaves your machine. On Team and Enterprise plans, optional cloud-synced test history is stored by TriageKit; see the Privacy Policy.
Do I need my own AI API key?
Yes, unless you run a local model. Add an API key from the AI provider you want to use. Usage is billed by that provider to your account and governed by your agreement with them.
Does TriageKit train models on my data?
No. TriageKit does not train models. How your AI provider handles API data is governed by that provider’s terms for your account.
Who can see my screenshots?
On your machine, only you. API requests go to the AI provider you chose, under your account with them. With a local model, nothing is sent out. On plans with cloud-synced history, members of your workspace can see the shared history your team chooses to sync.
Pricing and access
How do I get access?
TriageKit is in early access, and we’re onboarding teams in batches. Request access and we’ll reply from founder@triagekit.space.
What does it cost?
Solo is $19 per month, Team is $49 per seat per month, and Enterprise is custom. Prices are in USD. AI usage is billed separately by your AI provider; a local model has no provider bill. See Pricing.
Can I cancel any time?
Yes. Monthly plans can be cancelled at any time and stay active until the end of the current billing period.
Do you offer discounts for larger teams?
Enterprise plans are priced to fit. Email us with your team size and requirements.