2025/07/30 by Junde Wu, Wu, Junde, Minhao Hu +11 · 2 citations
Computer Science · #Software System Performance and Reliability #Advanced Software Engineering Methodologies #Software Engineering Research
paper · pdf · doi:10.48550/arxiv.2508.00031
Large language model (LLM) agents have demonstrated strong capabilities in long-horizon tasks by interleaving reasoning with tool use. However, as these agents scale to complex workflows such as software engineering and open-ended research, context management becomes a fundamental bottleneck: interaction histories grow unbounded, become costly to maintain, and are difficult to reuse across sessions and agents. We introduce Git-Context-Controller (GCC), a structured context management framework inspired by software version control systems. GCC elevates agent context from a transient token stream to a persistent, navigable memory workspace with explicit operations -- COMMIT, BRANCH, MERGE, and CONTEXT, that enable milestone-based checkpointing, isolated exploration of alternative reasoning paths, and hierarchical retrieval of historical context. By organizing agent memory as a versioned file system, GCC allows agents to manage long-term goals, recover and transfer reasoning across sessions, and coordinate multi-trajectory problem solving in a principled manner. Empirically, agents equipped with GCC achieve state-of-the-art performance on both SWE-Bench and BrowseComp benchmarks. On SWE-Bench Verified, GCC improves task resolution by over 13% relative to strong long-context baselines and outperforms 26 existing open and commercial systems, reaching over 80% success rate. The project will be open-sourced for the research community. https://github.com/ImprintLab/git-context-controller