RepoMate – AI Developer Assistant

Developers working with unfamiliar codebases often spend significant time understanding project structure, locating important files, reading documentation, and tracing existing code before they can make meaningful changes. This becomes especially challenging during onboarding, when a developer needs to quickly understand a project they did not build. Debugging creates a similar challenge because error messages often indicate that something went wrong without clearly explaining the underlying cause or how to approach a fix. RepoMate is an AI-powered developer workflow assistant designed to simplify both repository understanding and debugging. The application provides two main workflows. First, Understand a Repo allows a developer to enter a public GitHub repository URL. RepoMate retrieves the repository structure, README information, and a limited set of important source and configuration files. This context is organized and sent to an AI model, which produces a plain-English explanation of the project's purpose, structure, important files, and where a new developer should begin. Second, Explain a Bug allows a developer to enter an error message and optionally provide the related code. RepoMate sends this information to an AI model, which identifies the likely cause, explains the problem in simple terms, and suggests a practical approach to fixing it. The target users are students, new developers, contributors, and developers who need to quickly understand an unfamiliar project or investigate a coding issue. RepoMate is built using Node.js and Express with a lightweight web frontend. It uses Gemini for AI-generated explanations and includes Docker support for reproducible deployment. The application is publicly deployed so users can access it through a web browser. The creative aspect of RepoMate is its focus on providing context-aware developer explanations rather than simply returning generic AI answers. For repository analysis, relevant project context is collected first so the explanation is based on the actual codebase.