Can AI explain code on my screen?

Yes. Screen-aware AI explanation is one of the strongest applications of multimodal models. With Jarvis (getjarvis.eu), open any code — an IDE, a GitHub PR, a Stack Overflow answer, a screenshot, a PDF paper — press Cmd+/ on Mac or Ctrl+/ on Windows, and ask "explain this". Jarvis screenshots your current view, sends it to a multimodal model (frontier models from Anthropic, OpenAI, or Google), and returns a plain-English walkthrough of what the code does, why, and any potential issues. No copy-paste required. Persistent memory remembers your tech stack across sessions, so explanations stay relevant to your context. Free tier covers daily explanations; Pro at $16/month and Unlimited at $32/month raise quotas. Works on Mac, Windows, and Linux alongside Cursor, GitHub Copilot, and Continue — Jarvis covers the explanation use case those IDE assistants don't focus on.

Jarvis uses multimodal vision: screenshot capture + LLM. When you press Cmd+/ on Mac or Ctrl+/ on Windows, Jarvis captures the screen, sends the image (and your question) to frontier models from Anthropic, OpenAI, or Google, and the model reads the code directly from the pixels. This works even for code in places where text isn't easily extractable: PDFs of papers, screenshots from Twitter, embedded code in design tools, terminal output you can't easily copy. The explanation covers: what the code does line by line, the underlying concepts (recursion, async/await, decorators), why it's structured this way, edge cases or bugs you might miss, and suggested simpler alternatives where appropriate. Jarvis routes per task across frontier models from Anthropic, OpenAI, and Google (voice uses OpenAI realtime), switching to a frontier model for very dense code (hundreds of lines visible). Answers come back in seconds.

Install Jarvis (Mac, Windows, Linux). Sign in. Grant Screen Recording permission on first launch (revocable in System Settings). Now anytime you see code you want explained: press Cmd+/ on Mac or Ctrl+/ on Windows. Type "explain this" or "what does this code do?" or be specific: "why does this use a generator instead of a list?". Jarvis screenshots the screen, sends it to the model, returns an explanation in seconds. To go deeper: "explain it like I'm new to Python", "compare this approach to the one in the previous file", "what could go wrong with this?". You can also follow up: "now show me how to refactor it" — Jarvis remembers the code from the previous turn. Works in any IDE (VS Code, JetBrains, Cursor, Xcode, Vim), any browser, any terminal — anywhere code appears on your screen.

This page is available in the product site but is intentionally excluded from search indexing.

Code & tickets