TraceDev: A Traceability-Driven Multi-agent Framework for Requirement-to-Code Development

2026-07-21

AI Summary

This work addresses the challenge of ensuring consistency between natural language requirements and generated code, a limitation in existing approaches that rely on simplified instructions and lack explicit traceability mechanisms. To overcome this, the authors propose TraceDev, a multi-agent automated software development framework comprising five role-based agents—Requirements Refinement, Design, Development, Testing, and Validation—that collaboratively perform end-to-end code generation for complex use cases. The key innovation lies in the Validator agent, which constructs and maintains a heterogeneous traceability graph to explicitly link requirements, design models, and code, thereby enabling cross-artifact consistency assurance and structured contextual memory. Evaluated on the ETOUR and SMOS datasets, TraceDev achieves success rates of 53.63% and 56.82%, respectively, representing a maximum improvement of 340.80% over baseline methods and significantly outperforming current state-of-the-art approaches.

Abstract

In modern software development, the rapid advancement of Large Language Models (LLMs) has made the end-to-end transformation of Natural Language Requirements (NLRs) into executable repository-level code increasingly feasible. However, existing approaches typically rely on simplified instructions (e.g., single-sentence descriptions), failing to reflect complex software development scenarios. Moreover, they lack explicit requirement traceability mechanisms, making it difficult to precisely align and validate generated code against original requirements. To address these limitations, we propose TraceDev, a multi-agent framework for automated software development grounded in use cases that contain multiple functional points and complex semantics. TraceDev employs five role-specific agents, including a Requirement Refiner, Designer, Developer, Tester, and Validator. Notably, the Validator Agent constructs and maintains a heterogeneous traceability graph that links requirements, design models, and code artifacts for interacting with the preceding four agents. The traceability graph maintains consistency across various artifacts and serves as a structured context for efficient memory management, supporting reliable repository-level code generation. We evaluate TraceDev on two widely used datasets (including 125 use cases) compared with two state-of-the-art approaches. On the ETOUR dataset, TraceDev achieves a success rate of 53.63\%, outperforming baseline approaches by up to 186.63\%. A similar trend is observed on the SMOS dataset, where TraceDev attains a success rate of 56.82\%, exceeding baseline approaches by up to 340.80\%. These results demonstrate the effectiveness of TraceDev in repository-level code generation from requirements.

Problem

• requirement-to-code
• traceability
• natural language requirements
• repository-level code generation
• software development

Innovation

• traceability graph
• multi-agent framework
• requirement-to-code generation
• Large Language Models
• repository-level code