RepoLens: Smarter Repository Understanding

Understanding an unfamiliar software repository can take hours, especially for students, new developers, interns, and engineers joining an existing project. Developers often need to manually navigate through files and folders, identify programming languages and frameworks, understand dependencies and configuration files, locate APIs and entry points, and figure out the overall architecture. Important information is scattered throughout the codebase, making onboarding slow and making specific questions difficult to answer quickly. Traditional code search can find files or keywords, but it does not provide a complete understanding of how the repository works. RepoLens addresses this problem by providing an AI-powered GitHub repository understanding and developer onboarding platform. A user simply provides a public GitHub repository URL, and RepoLens ingests and analyzes the repository to create an interactive, queryable understanding of the codebase. The platform presents repository metadata, programming languages, frameworks, architecture information, APIs and routes, dependencies, important files, and an interactive file explorer in one place. The core feature is Ask the Repository, which allows users to ask natural-language questions about the codebase. RepoLens uses a Retrieval-Augmented Generation (RAG) pipeline to provide repository-specific answers rather than generic AI responses. Source files are processed into code chunks while preserving their file paths and line numbers. When a user asks a question, the system retrieves the most relevant code context and provides it to a configurable language model. The generated response is grounded in the retrieved repository content and includes file and line-level citations, allowing developers to verify the answer directly in the source code. The primary target users are developers, students, interns, software teams, and anyone who needs to understand an unfamiliar codebase quickly. Users interact with RepoLens through a simple web interface: they submit a GitHub repository URL, explore the generated repository analysis, inspect architecture and dependencies, navigate the file structure, and ask questions through the AI assistant. What makes RepoLens distinctive is the combination of repository analysis, code exploration, RAG-based AI, and source-level traceability in a single workflow. Instead of treating AI as a generic chatbot, RepoLens connects responses directly to the repository's actual source code. This makes the system more useful for technical understanding and verification. RepoLens improves efficiency by reducing repetitive manual searching and helping developers move quickly from an unfamiliar repository to a structured understanding of its code, architecture, and implementation. The project was developed using IBM Bob across the workflow, including planning, implementation, repository ingestion, analysis, RAG, LLM integration, testing, and final polishing.