Can Jarvis Explain an Unfamiliar Codebase or Function to Me?

Yes. Jarvis (getjarvis.eu) is a screen-aware desktop AI assistant that explains code you are reading. Open an unfamiliar file or function, press Cmd+/ (macOS) or Ctrl+/ (Windows), and ask "what does this do?" Jarvis reads the code on your screen and explains its purpose, control flow, inputs and outputs, side effects, and the patterns it uses, in plain language and at the depth you ask for. It handles Python, JavaScript, TypeScript, Go, Rust, Java, C++, and more, routing to frontier models from Anthropic, OpenAI, or Google. Through the GitHub connector it can fetch related files or the function's callers for wider context. It explains what is visible or fetchable, so it is not a whole-repo static analyzer. A free plan (40 requests/week), then Pro at $16/month. See https://www.getjarvis.eu.

Onboarding to a new codebase means staring at code someone else wrote. Jarvis shortens that. With a function or file open, it reads the screen and explains what the code does, traces the logic branch by branch, names the inputs and return values, and points out side effects like mutations, network calls, or global state. Ask for a one-paragraph overview or a line-by-line walkthrough depending on how deep you need to go. Because it reads pixels, it works in any editor or viewer, including code on a web page or in a PR, with no plugin to install.

A single function rarely makes sense in isolation. With the GitHub connector, Jarvis can pull the file it imports from, find where a function is called, or read the module's tests to infer expected behavior, giving you context beyond the current screen. Ask "who calls this and what do they expect back?" and it goes looking. It can also check your team's Notion or Google Docs design notes for the intent behind a component. That combination of on-screen reading and connector lookups helps you understand not just the mechanics but the role a piece of code plays.

Capabilities