The paper investigates how coding agents handle large repositories when their context window is limited. It compares three strategies—prompt compression, code retrieval, and a simple file selector called CQ2-CARE—finding that allocating the token budget to selecting whole files verbatim enables the agent to fix more issues. CQ2-CARE outperforms both compressors and retrievers on a benchmark of 300 recent software‑engineering problems, achieving higher success rates with GPT‑5.6 and Claude Opus 5.
arXiv:2607. 24882v1 Announce Type: cross Abstract: Modern coding agents are usually evaluated by whether they eventually produce a correct patch, but patch generation depends on an earlier context-acquisition stage: finding the repository files needed for the task.
Modern LLM coding agents such as Claude Code and OpenHands share a common inefficiency: they spend much of their token budget finding the file to patch, rather than patching it. On SWE-Bench Verified, a 30B OpenHands agent averages 23 rounds and 631K tokens per resolved issue, with many calls spent on grep, glob, and view_file during repository exploration.
arXiv:2608. 05886v1 Announce Type: cross Abstract: Modern LLM coding agents such as Claude Code and OpenHands share a common inefficiency: they spend much of their token budget finding the file to patch, rather than patching it.
arXiv:2607. 19653v1 Announce Type: cross Abstract: Large language model (LLM) agents now perform well on correctness-oriented repository-level tasks, including SWE-Bench issue resolution and feature implementation in real codebases.
arXiv:2607. 09691v1 Announce Type: cross Abstract: A modern coding agent can hold an entire repository in its context window.
arXiv:2608.29675v1 Announce Type: cross Abstract: Repository exploration is a distinct and costly stage of coding-agent pipelines: before generating a patch, an agent must identify which repository f...
The paper explores whether natural‑language documentation aids coding agents in fixing software bugs and introduces a roundtrip benchmark that evaluates code descriptions by regenerating code and testing it. It finds that description completeness, not length, determines fidelity, and presents an optimizer that can produce fully faithful descriptions that generalize to new files. However, experiments across two model families and ten repositories show that such compact documentation does not improve an agent’s ability to resolve real repository issues compared to using the issue alone.
The paper introduces Stratum, a repository memory system designed to give coding agents persistent, experience-aware memory beyond the limits of their context windows. It argues that coding agents suffer from a memory-architecture problem, not just context length, and presents three empirical findings from real repositories that motivate a tiered memory approach. Stratum stores knowledge as indexed text rather than in the model’s KV cache, enabling fast retrieval (1.7 ms median for a million nodes) and significantly improving root-cause identification in multi‑session bug‑fix experiments.
arXiv:2607. 18161v1 Announce Type: cross Abstract: Coding agents are increasingly used to accelerate code generation in many downstream tasks, such as fixing bugs, building applications, and prototyping.
Context windows have grown a thousand-fold, yet coding agents still start every session with no memory of their past work. They re-read files, retry failed approaches, and forget root causes they esta...
Paritok-4B is a 4‑billion‑parameter LoRA compressor designed for coding agents, which extracts and retains key spans of code rather than paraphrasing them. It is intent‑conditioned, selecting lines that are most relevant to the agent’s current task, and achieves high fidelity with 96% of identifiers, paths, and numbers preserved. Trained on 67,074 real OpenHands trajectories and fine‑tuned on Qwen3‑4B, it compresses agent context to about 25.7% of its original size while keeping 86.5% of the uncompressed solve quality across 300 SWE‑bench Lite instances.
Coding agents re-send large file reads and tool outputs to a frontier LLM every turn, and this context dominates their token bill. General-purpose prompt compressors are trained on prose and suit code...