RepoNav helps AI find the right code files and functions faster
RepoNav: From Snippet Retrieval to File-Centered Repository Navigation for Code Agents
Software EngineeringComputation and Language
Summary
Finding the right piece of code in a big project can be hard, especially when many parts look similar. The authors created RepoNav, a tool that organizes search results by showing code snippets grouped by the files they come from. This helps AI agents better explore and compare related functions in the same file before choosing the right one. Their tests showed that this structured way of presenting code improves how well AI can locate specific functions and answer questions about code projects.
large language modelscode searchcode snippetsrepository navigationfunction localizationsoftware engineeringcode agentsstructured retrieval
Authors
Hongzheng Chai, Jiakun Li, Hongyue Yu, Yuan Yuan
Abstract
Solving repository-level code tasks requires LLM-based agents to use code search tools to navigate large codebases and identify a small set of relevant files and functions. However, current retrieval tools typically return flat lists of isolated code snippets: such lists can surface relevant files, but provide insufficient structure for agents to distinguish the target function from semantically similar alternatives in the same file. We introduce RepoNav, a lightweight post-retrieval interface that reorganizes retrieved snippets into a file-centered navigation scaffold. By presenting compact structural cues and candidate targets, this scaffold guides on-demand file-structure browsing, helping agents compare sibling symbols before selecting a target function. Across diverse models on LocBench, RepoNav improves function-level localization and narrows the file-to-function gap. Controlled ablations demonstrate that these gains come from structured evidence organization rather than simply exposing additional file structure, and the approach also improves performance on a repository-level question-answering benchmark.