DocAgent: A Multi-Agent System for Automated Code Documentation Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yang, Dayu, Simoulin, Antoine, Qian, Xin, Liu, Xiaoyi, Cao, Yuwei, Teng, Zhaopu, Yang, Grey
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909621797519360
author Yang, Dayu
Simoulin, Antoine
Qian, Xin
Liu, Xiaoyi
Cao, Yuwei
Teng, Zhaopu
Yang, Grey
author_facet Yang, Dayu
Simoulin, Antoine
Qian, Xin
Liu, Xiaoyi
Cao, Yuwei
Teng, Zhaopu
Yang, Grey
contents High-quality code documentation is crucial for software development especially in the era of AI. However, generating it automatically using Large Language Models (LLMs) remains challenging, as existing approaches often produce incomplete, unhelpful, or factually incorrect outputs. We introduce DocAgent, a novel multi-agent collaborative system using topological code processing for incremental context building. Specialized agents (Reader, Searcher, Writer, Verifier, Orchestrator) then collaboratively generate documentation. We also propose a multi-faceted evaluation framework assessing Completeness, Helpfulness, and Truthfulness. Comprehensive experiments show DocAgent significantly outperforms baselines consistently. Our ablation study confirms the vital role of the topological processing order. DocAgent offers a robust approach for reliable code documentation generation in complex and proprietary repositories.
format Preprint
id arxiv_https___arxiv_org_abs_2504_08725
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DocAgent: A Multi-Agent System for Automated Code Documentation Generation
Yang, Dayu
Simoulin, Antoine
Qian, Xin
Liu, Xiaoyi
Cao, Yuwei
Teng, Zhaopu
Yang, Grey
Software Engineering
Artificial Intelligence
Computation and Language
Machine Learning
High-quality code documentation is crucial for software development especially in the era of AI. However, generating it automatically using Large Language Models (LLMs) remains challenging, as existing approaches often produce incomplete, unhelpful, or factually incorrect outputs. We introduce DocAgent, a novel multi-agent collaborative system using topological code processing for incremental context building. Specialized agents (Reader, Searcher, Writer, Verifier, Orchestrator) then collaboratively generate documentation. We also propose a multi-faceted evaluation framework assessing Completeness, Helpfulness, and Truthfulness. Comprehensive experiments show DocAgent significantly outperforms baselines consistently. Our ablation study confirms the vital role of the topological processing order. DocAgent offers a robust approach for reliable code documentation generation in complex and proprietary repositories.
title DocAgent: A Multi-Agent System for Automated Code Documentation Generation
topic Software Engineering
Artificial Intelligence
Computation and Language
Machine Learning
url https://arxiv.org/abs/2504.08725