Context as a Tool: Context Management for Long-Horizon SWE-Agents

2025-12-26

AI Summary

To address semantic drift, reasoning degradation, and context explosion in long-horizon software engineering agents operating over ultra-large-scale codebases—stemming from uncontrolled context growth—this paper introduces the “Context-as-Tool” (CAT) paradigm, which explicitly models context management as callable, learnable tools. Methodologically: (1) we construct a structured workspace that decouples high-fidelity short-term interactions from compressed long-term memory; (2) we design CAT-GENERATOR, a trajectory-level supervised framework enabling milestone-driven proactive compression; and (3) we develop SWE-Compressor, a context-aware compression model. Evaluated on SWE-Bench-Verified, our approach achieves a 57.6% task success rate—significantly outperforming ReAct baselines and static compression methods—while ensuring robustness and scalability of long-range reasoning under fixed context budgets.

Technology Category

• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories

Application Category

• Search and Retrieval-Augmented AI: Large language models for search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work

Abstract

Agents based on large language models have recently shown strong potential on real-world software engineering (SWE) tasks that require long-horizon interaction with repository-scale codebases. However, most existing agents rely on append-only context maintenance or passively triggered compression heuristics, which often lead to context explosion, semantic drift, and degraded reasoning in long-running interactions. We propose CAT, a new context management paradigm that elevates context maintenance to a callable tool integrated into the decision-making process of agents. CAT formalizes a structured context workspace consisting of stable task semantics, condensed long-term memory, and high-fidelity short-term interactions, and enables agents to proactively compress historical trajectories into actionable summaries at appropriate milestones. To support context management for SWE-agents, we propose a trajectory-level supervision framework, CAT-GENERATOR, based on an offline data construction pipeline that injects context-management actions into complete interaction trajectories. Using this framework, we train a context-aware model, SWE-Compressor. Experiments on SWE-Bench-Verified demonstrate that SWE-Compressor reaches a 57.6% solved rate and significantly outperforms ReAct-based agents and static compression baselines, while maintaining stable and scalable long-horizon reasoning under a bounded context budget.

Problem

• Addresses context explosion in long-horizon software engineering agents
• Proposes proactive context compression to maintain stable reasoning
• Enables scalable agent interactions with repository-scale codebases

Innovation

• Context management as callable tool for agents
• Structured workspace with proactive compression at milestones
• Offline supervision framework trains context-aware compressor model