Context management, not planning, drives coding agent performance under tight budgets

An Empirical Study of Harness Design for Coding Agents

A 43-page empirical study isolates three harness components—planning, action space, and context management—across 176 matched settings on SWE-Bench Verified and Terminal-Bench 2.1 with four models. Context management matters most as the context-window budget tightens, mainly by preventing overflow failures. Rule-based elision before LLM summarization is the most efficient strategy; making elided content recoverable adds unused machinery. Planning shifts from an accuracy scaffold for weaker models to a cost saver for stronger ones. Predefined tools help models with weaker bash skills, while bash-capable models do fine with a bash-only interface at lower cost.