🤖 AI Summary
Existing automated code review approaches often suffer from diminished relevance, increased redundancy, and ambiguous prioritization in repository-scale settings due to the conflation of multidimensional tasks. This work proposes a locality-first, modular multi-agent architecture that decomposes the review process into five distinct stages: repository ingestion, context synthesis, file-level analysis, issue prioritization, and summary generation. Implemented with Python CLI, FastAPI, LangGraph, and Next.js, the system delivers end-to-end localized code review capabilities and includes reusable infrastructure for evaluation and reporting. Beyond presenting a practical system design, this study explicitly articulates key architectural trade-offs, developer-facing interfaces, and common failure modes, thereby establishing a foundational platform for future empirical research in automated code review.
📝 Abstract
Repository-level code review requires reasoning over project structure, repository context, and file-level implementation details. Existing automated review workflows often collapse these tasks into a single pass, which can reduce relevance, increase duplication, and weaken prioritization. We present RepoReviewer, a local-first multi-agent system for automated GitHub repository review with a Python CLI, FastAPI API, LangGraph orchestration layer, and Next.js user interface. RepoReviewer decomposes review into repository acquisition, context synthesis, file-level analysis, finding prioritization, and summary generation. We describe the system design, implementation tradeoffs, developer-facing interfaces, and practical failure modes. Rather than claiming benchmark superiority, we frame RepoReviewer as a technical systems contribution: a pragmatic architecture for repository-level automated review, accompanied by reusable evaluation and reporting infrastructure for future empirical study.