An Empirical Study of Harness Design for Coding Agents

arXiv:2609.20804 · cs.AI, cs.CL, cs.LG, cs.SE · Submitted 2026-09-17 · Read on arXiv

cs.AI, cs.CL, cs.LG, cs.SE

Submitted: 2026-09-17

Updated: 2026-09-17

Comments: 43 pages

Code: https://github.com/anomalyco/opencode

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Action Space — 25× in this paper · explained in 1 episode
• Benchmark — 25× in this paper · explained in 2 episodes
• Accuracy — 20× in this paper · explained in 1 episode
• Verification — 18× in this paper · explained in 2 episodes
• Ablation — 13× in this paper · explained in 1 episode
• Scaffolding — 5× in this paper · explained in 3 episodes
• LLM judge — 4× in this paper · explained in 2 episodes
• Commentary — 3× in this paper · explained in 2 episodes
• Agent Architectures — 2× in this paper · explained in 1 episode
• Core — 2× in this paper · explained in 2 episodes
• Trace — 2× in this paper · explained in 2 episodes
• AI Agents — 1× in this paper · explained in 2 episodes
• Agent Harness — 1× in this paper · explained in 1 episode
• Agentic Task — 1× in this paper · explained in 1 episode
• Agentic Tasks — 1× in this paper · explained in 1 episode
• Alignment — 1× in this paper · explained in 1 episode
• Benchmarking — 1× in this paper · explained in 3 episodes
• Large Language Models (LLMs) — 1× in this paper · explained in 6 episodes
• Modular Architecture — 1× in this paper · explained in 4 episodes
• Prompt Engineering — 1× in this paper · explained in 4 episodes
• Reinforcement Learning — 1× in this paper · explained in 4 episodes
• Task Planning — 1× in this paper · explained in 2 episodes
• Verifier — 1× in this paper · explained in 2 episodes

Sources

• MASAI: Modular Architecture for Software-engineering AI Agents
• NVIDIA Nemotron 3: Efficient and Open Intelligence
• AgentArch: A Comprehensive Benchmark to Evaluate Agent Architectures in Enterprise
• Qwen3-Coder-Next Technical Report
• CodeR: Issue Resolving with Multi-Agent and Task Graphs
• LCM: Lossless Context Management
• ACON: Optimizing Context Compression for Long-horizon LLM Agents
• Same Model, Different Harness: Different Coding-Agent Results
• More Is Not Always Better: Cross-Component Interference in LLM Agent Scaffolding
• Beyond Resolution Rates: Behavioral Drivers of Coding Agent Success and Failure
• MemGPT: Towards LLMs as Operating Systems
• Inside the Scaffold: A Source-Code Taxonomy of Coding Agent Architectures
• Kimi K2.5: Visual Agentic Intelligence
• Executable Code Actions Elicit Better LLM Agents
• Confucius Code Agent: Scalable Agent Scaffolding for Real-World Codebases
• Recursively Summarizing Books with Human Feedback
• ReSum: Unlocking Long-Horizon Search Intelligence via Context Summarization
• Agentless: Demystifying LLM-based Software Engineering Agents
• LLM Agents Are Latent Context Managers: Eliciting Self-Managed Context via State Proprioception
• ReAct: Synergizing Reasoning and Acting in Language Models

Related papers

• MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
• Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
• The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
• MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
• Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems
• VSAL: A Vision Solver with Adaptive Layouts for Graph Property Detection