Super Library Agent: Joint Generation and Maintenance of Multiple Applications Beyond the Single Codebase
Software EngineeringArtificial IntelligenceComputation and Language
Summary
The authors study how AI coding assistants can build and update multiple related software applications that share common parts. Instead of copying shared code for each app, they want the AI to create a single shared library to reduce repeated work and keep things cleaner. They find that a simple step-by-step approach to making this shared library doesn't work well, so they improve it using ways to identify shared code better and track dependencies more carefully. Their method keeps the apps working correctly while making the code shorter and less duplicated compared to simpler approaches.
Authors
Daegyu Sung, Yukyeong Lee, Geon Park, Yumin Choi, Sung Ju Hwang
Abstract
Organizations often develop and maintain portfolios of related applications: independently deployable codebases that share substantial domain logic, interface patterns, or operational conventions. As LLM coding agents are increasingly used to generate and maintain such software, a naive application-by-application workflow duplicates shared logic across codebases and allows prolonged agentic maintenance to accumulate verbosity, dead code, and structural erosion. We introduce the Super Library Agent problem, where an agent sequentially generates a portfolio of N related applications while maintaining a shared Super Library of reusable cross-application components. A minimal sequential scaffold can in principle extract shared code and migrate applications to the evolving library, but in practice suffers from low extraction recall and fragile dependency migration. We address these failures with candidate-guided extraction over code chunk summaries, pre-extraction codebase consolidation, and context-aware migration using extraction traces and call-graph information. Across WebGen-Bench and PaperBench, our method preserves application functionality while significantly reducing redundancy and token footprint (verbosity, token length) over zero-shot, and avoiding the structural erosion introduced by naive library construction, with additional reductions in LOC and MDL. Our code is available at https://github.com/sbigstar0310/super-library-agent.