AI summary · arXiv: 2604.25850v4 · cs.CL
Agentic Harness Engineering: Observability-Driven Automatic Evolution of Coding-Agent Harnesses
The essentials
- Coding-agent performance plays a pivotal role in facilitating interactions between models and tools.
- Challenges in coding-agent performance stem from diverse action space, extensive trajectories, and difficult-to-trace edits.
- Agentic Harness Engineering (AHE) introduces a closed-loop system with three observability pillars: component observability, experience observability, and decision observability.
- AHE transforms edits into falsifiable contracts to improve performance without trial-and-error methods.
- Empirical results show that AHE significantly improves pass@1 performance on Terminal-Bench 2 from 69.7% to 77.0%, surpassing human-designed harness Codex-CLI and self-evolving baselines ACE and TF-GRPO.
- The evolved harness can be transferred without re-evolution, achieving superior success rates with fewer tokens than the initial seed across different model families on Terminal-Bench 2.
- Ablation studies attribute improvement to tools, middleware, and long-term memory rather than system prompts.
Key points written by AI: they may contain errors, check them against the paper.
Ask questions about this paper to our AI assistant
How to chat with this paper? Illustrated guide
You can also chat with multiple papers at once here.
Keep this conversation — a free account saves your questions and answers, your collections of papers and your alerts.
Create a free accountTools for this paper
Advanced tools Premium
For researchers who follow a field closely: citation alerts, journal-club slides, version comparison and Zotero export. Discover Premium
Papers around this one
See them on a mapRanked by how close their title and abstract are to this paper’s. Select some to read them together.
Closest papers
-
HarnessX: A Composable, Adaptive, and Evolvable Agent Harness FoundryTing-Yang Chen, Shuo Lu, Kangjia Zhao et al. · 2026 · arXiv.org · preprint, not peer reviewed · cited 60 times · Not summarized yet
Abstract
AI agent performance depends critically on the runtime harness, comprising the prompts, tools, memory, and control flow that mediate how a model observes, reasons, and acts. Yet today's harnesses remain largely hand-crafted and static: each new model or task still demands bespoke scaffolding, and the rich traces produced during execution are rarely distilled back into systematic improvement. We introduce HarnessX, a foundry for composable, adaptive, and evolvable agent harnesses. HarnessX assembles typed harness primitives via a substitution algebra, adapts them through AEGIS, a trace-driven multi-agent evolution engine grounded in an operational mirror between symbolic adaptation and reinfo…
-
HarnessCompass: Guiding Automatic Harness Evolution toward Generalizable and Effective Agent HarnessesLuan Zhang, Ruo-Chen Zhou, Dan-Dan Song et al. · 2026 · cited 24 times · Not summarized yet
Abstract
Harness design plays a critical role in agent performance by shaping how large language models (LLMs) perceive, reason over, and act within executable environments. Recent work has proposed automatic harness evolution, which iteratively improves the harness from agent--environment interactions. However, existing methods often overfit to the evolution tasks, rely exclusively on trajectory-derived signals, and optimize harness components jointly, causing interference across components. We propose HarnessCompass, a novel automatic harness evolution framework built around constrained evolution, proactive feedback, and component-wise optimization. HarnessCompass first enforces global constraints …
-
Self-Evolving Harness on Multiple Tasks with the Agent as Its Own OptimizerQian-Kai Xu · 2026 · cited 1 time · Not summarized yet
Abstract
A harness is the code around a language-model agent that organizes prompts, calls tools, manages context, and controls execution. As models grow stronger, recent work has begun to let agents improve their own harnesses, a line of work known as self-evolving harnesses. In most existing methods, a separate proposer running on a human-designed harness modifies the solver's harness, and a separate harness is evolved for each benchmark. Real-world tasks come from many domains, so both the evolution and the evaluation of a harness should cover a diverse range of tasks. We propose a framework close to recursive self-improvement: the same frozen model, on the same version of the harness, first solve…
-
Adaptive Auto-Harness: Sustained Self-Improvement for Agentic System Deployment on Open-Ended Task StreamsZe-Wen Liu, Zhan Shi, Yi-Si Sang et al. · 2026 · arXiv.org · preprint, not peer reviewed · cited 33 times · Not summarized yet
Abstract
Auto-harness systems such as A-Evolve, GEPA, and Meta-Harness improve LLM agents by optimizing prompts, skills, tools, memories, and supporting infrastructure from execution feedback, but they are typically evaluated on fixed offline benchmarks. Real deployments instead present open-ended task streams: histories grow without a fixed endpoint, heterogeneous tasks require different harnesses, and problem distributions shift over time. These challenges make a single repeatedly and densely updated harness brittle, causing performance degradation as accuracy peaks early and then declines. This motivates sustained harness construction with task-wise adaptation. We introduce Adaptive Auto-Harness, …
-
Self-Harness: Harnesses That Improve ThemselvesHangfan Zhang, Shao Zhang, Kangcong Li et al. · 2026 · Computation and Language · preprint, not peer reviewed · cited 123 times · AI summary available
Self-Harness studies agents improving their own operating harnesses through an iterative loop, as AHE does, though AHE centers observability and falsifiable edits.
Abstract
The performance of LLM-based agents is jointly shaped by their base models and the harnesses that mediate their interaction with the environment. Because different models exhibit distinct behaviors, effective harness design is inherently model-specific. Yet agent harnesses are still largely engineered by human experts, a paradigm that scales poorly as modern LLMs become increasingly diverse and rapidly evolving. In this paper, we introduce Self-Harness, a new paradigm in which an LLM-based agent improves its own operating harness, without relying on human engineers or stronger external agents. We operationalize Self-Harness as an iterative loop with three stages: Weakness Mining, which ident…
-
Harness Handbook: Making Evolving Agent Harnesses Readable,Navigable, and EditableRu-Han Wang, Yu-Cheng Shi, Zong-Xia Li et al. · 2026 · arXiv.org · preprint, not peer reviewed · cited 15 times · Not summarized yet
Abstract
The capability of a modern AI agent depends not only on its foundation model but also on its harness, which constructs prompts, manages state, invokes tools, and coordinates execution. As models, APIs, environments, and requirements evolve, the harness must be continually modified. Before such a change can be made, a developer or coding agent must identify all code locations that implement the target behavior. This is difficult because production harnesses are large, tightly coupled, and behaviorally distributed, while modification requests describe what the system should do and repositories are organized by files and modules. Code search, repository indexing, and long-context processing eas…
-
HarnessDev: Can LLMs Create and Evolve Their Own Agent Harness?Yu-Hao Wu, Jing-Yuan Zhang, Jia-Jun Shi et al. · 2026 · cited 15 times · Not summarized yet
Abstract
As agents move from research prototypes to deployed tools, their capability increasingly depends on model-external execution infrastructure, commonly termed the agent harness. Changing this harness while holding model weights fixed can substantially alter task performance. Current agent evaluations typically report downstream performance under a chosen harness, leaving a model's ability to develop the harness itself comparatively underexplored. We introduce HarnessDev, a benchmark that shifts the unit of evaluation from task outputs to runnable infrastructure. HarnessDev covers two stages. In Creation, the agent starts from a minimal seed and a small number of cases, then builds a complete e…
-
Causal Improvement Graph for Agentic Harness OptimizationJun-Jie Zhang, Shunyu Liu, Hao-Yu Wang et al. · 2026 · Not summarized yet
Abstract
Agentic Harness is the runtime that constructs task context and controls execution flow, thereby shaping overall agent performance. Given a fixed model and external evaluation, automated Harness optimization seeks to improve this runtime through an iterative proposal--evaluation loop to better solve target tasks. Existing meta-harness methods mainly adopt proposer-centric discovery, in which an LLM-based proposer integrates accumulated experimental findings to determine subsequent Harness revisions. This places the burden of maintaining the evolving improvement state on the proposer as history expands and its underlying experimental logic becomes harder to discern. In this paper, we introduc…
Prior work it builds on
-
Meta-Harness: End-to-End Optimization of Model HarnessesYoonho Lee, R. Nair, Qizheng Zhang et al. · 2026 · arXiv.org · preprint, not peer reviewed · cited 353 times · Not summarized yet
Abstract
The performance of large language model (LLM) systems depends not only on model weights, but also on their harness: the code that determines what information to store, retrieve, and present to the model. Yet harnesses are still designed largely by hand, and existing text optimizers are poorly matched to this setting because they compress feedback too aggressively. We introduce Meta-Harness, an outer-loop system that searches over harness code for LLM applications. It uses an agentic proposer that accesses the source code, scores, and execution traces of all prior candidates through a filesystem. On online text classification, Meta-Harness improves over a state-of-the-art context management s…
-
Agentic Context Engineering: Evolving Contexts for Self-Improving Language ModelsQizheng Zhang, Chang-Ran Hu, Shubhangi Upasani et al. · 2025 · arXiv.org · preprint, not peer reviewed · cited 500 times · Not summarized yet
Abstract
Large language model (LLM) applications such as agents and domain-specific reasoning increasingly rely on context adaptation: modifying inputs with instructions, strategies, or evidence, rather than weight updates. Prior approaches improve usability but often suffer from brevity bias, which drops domain insights for concise summaries, and from context collapse, where iterative rewriting erodes details over time. We introduce ACE (Agentic Context Engineering), a framework that treats contexts as evolving playbooks that accumulate, refine, and organize strategies through a modular process of generation, reflection, and curation. ACE prevents collapse with structured, incremental updates that p…
-
Automated Design of Agentic SystemsSheng-Ran Hu, Cong Lu, Jeff Clune · 2024 · International Conference on Learning Representations · cited 494 times · Not summarized yet
Abstract
Researchers are investing substantial effort in developing powerful general-purpose agents, wherein Foundation Models are used as modules within agentic systems (e.g. Chain-of-Thought, Self-Reflection, Toolformer). However, the history of machine learning teaches us that hand-designed solutions are eventually replaced by learned solutions. We describe a newly forming research area, Automated Design of Agentic Systems (ADAS), which aims to automatically create powerful agentic system designs, including inventing novel building blocks and/or combining them in new ways. We further demonstrate that there is an unexplored yet promising approach within ADAS where agents can be defined in code and …
-
SWE-Hub: A Unified Production System for Scalable, Executable Software Engineering TasksYu-Chen Zeng, Shu-Peng Li, Da-Xiang Dong et al. · 2026 · arXiv.org · preprint, not peer reviewed · cited 3 times · Not summarized yet
Abstract
Progress in software-engineering agents is increasingly constrained by the scarcity of executable, scalable, and realistic data for training and evaluation. This scarcity stems from three fundamental challenges in existing pipelines: environments are brittle and difficult to reproduce across languages; synthesizing realistic, system-level bugs at scale is computationally expensive; and existing data predominantly consists of short-horizon repairs, failing to capture long-horizon competencies like architectural consistency. We introduce \textbf{SWE-Hub}, an end-to-end system that operationalizes the data factory abstraction by unifying environment automation, scalable synthesis, and diverse t…
-
SkillClaw: Let Skills Evolve Collectively with Agentic EvolverZi-Yu Ma, Shidong Yang, Yuxiang Ji et al. · 2026 · arXiv.org · preprint, not peer reviewed · cited 77 times · Not summarized yet
Abstract
Large language model (LLM) agents such as OpenClaw rely on reusable skills to perform complex tasks, yet these skills remain largely static after deployment. As a result, similar workflows, tool usage patterns, and failure modes are repeatedly rediscovered across users, preventing the system from improving with experience. While interactions from different users provide complementary signals about when a skill works or fails, existing systems lack a mechanism to convert such heterogeneous experiences into reliable skill updates. To address these issues, we present SkillClaw, a framework for collective skill evolution in multi-user agent ecosystems, which treats cross-user and over-time inter…
Papers that cite it
-
Rethinking the Evaluation of Harness Evolution for AgentsYike Wang, Huai-Sheng Zhu, Zheng-Yu Hu et al. · 2026 · arXiv.org · preprint, not peer reviewed · cited 62 times · Not summarized yet
Abstract
Harness evolution is an iterative search procedure that repeatedly evaluates and revises candidate harnesses used for LLM agents using task feedback. We revisit the evaluation of such automatic harness evolution procedures and identify two fundamental issues in the protocol. First, prior work does not compare these approaches with simple task-level search baselines under matched feedback and inference budgets. Second, prior work searches for harness configurations using verification signals (e.g., unit test cases) drawn from the same benchmarks on which it reports the final performance of the evolved harnesses, violating the standard separation between training and test data. To address this…
-
Harness-Delta Attribution: Dissecting Gains From Harness EvolutionWen-Xuan Ding, Qin Lu, Chang-Long Yu et al. · 2026 · Not summarized yet
Abstract
Harness evolution automatically optimizes the prompts, skills, tools, and orchestration code around a fixed-weight model, offering a way to improve agentic systems without updating model weights. The agent harness is iteratively optimized to achieve higher performance on the search set. But how much of this gain reflects benchmark-specific overfitting or additional inference-time compute, and how much may generalize beyond the search set? In this work, we introduce Harness-Delta Attribution (HDA), an evaluation protocol that dissects an evolution gain into artifact-driven overfitting, gains explained by test-time scaling, and the remainder, which gives an upper bound on residual generalizabl…
-
Hierarchical Self-Improvement: A Framework for Task-Specific Evolvable Agent HarnessesTailin Zhou · 2026 · cited 9 times · Not summarized yet
Abstract
Modern LLM agents are often improved by modifying prompts, tools, or workflows manually, while the executable scaffold surrounding the model---the \emph{harness}---is typically treated as a fixed artifact after deployment. This work studies an alternative where the harness is \emph{task-specific and continuously evolvable}: each task family maintains its own harness, which is hot-swapped across iterations through a fixed task-injection seam and rewritten using environment feedback. We introduce \textbf{Hierarchical Self-Improvement (HSI)}, a framework in which a single frozen LLM $M$ operates across three hierarchical scopes: a task harness $H$ that executes tasks, an evolver that rewrites $…
-
Harness-R1: Learning to Edit Executable Runtime Harnesses from Agent Failure TrajectoriesShuai Shao, Kang-Ning Zhang, Qingyao Li et al. · 2026 · cited 19 times · Not summarized yet
Abstract
Agents built around large language models continually accumulate interaction trajectories during deployment, yet their behavior typically remains fixed. Beyond updating model weights, these trajectories can improve the agent harness that constructs context, mediates tools, validates actions, and recovers execution. We introduce Harness-R1, the first method, to our knowledge, that makes failure-conditioned, lifecycle-wide editing of an existing executable runtime a learned capability. It post-trains a dedicated harness engineer with online reinforcement learning so that its edits are optimized for the realized task success they produce, rather than proposed by a fixed editor. A separate 9B en…
-
Verify Smarter, Evolve Further: Efficient Harness Evolution through Behavior-Aware VerificationJing-Heng Xu, Yi-Kai Zhang, Aiden Chen et al. · 2026 · cited 7 times · Not summarized yet
Abstract
Agent harnesses shape how language-model agents use instructions, tools, and runtime components, but adapting these harnesses requires costly verification. Existing propose-and-verify methods typically score every candidate on a fixed task set, wasting rollouts on unrelated behaviors and allowing aggregate scores to obscure specific regressions. We introduce HarnessLens, a budget-aware framework for automated harness evolution. HarnessLens jointly explores the task space and user-configurable components, derives candidate modifications from execution trajectories, and selectively verifies each candidate on behavior-relevant tasks using an attributable-evidence gate. Across three agent harnes…
Go further
How are these papers found?
We compare the title and abstract of this paper with those of every paper summarized on this site, and with the papers Semantic Scholar recommends, those it cites and those citing it (for a much-cited paper, its most recent citations and, from OpenAlex, the most cited ones). Among the prior work and the follow-ups, the papers that use each other in depth, a method it uses and the much-cited follow-ups come a little higher. The levels (very close, close, related) come from the similarity of the texts, not from a judgment of quality; * marks a paper without an abstract, compared by its title. The sentence under a paper is written by an AI from the two abstracts: check it in the papers. A paper marked not relevant by three readers leaves the list.
Updated on 31 October 2026.
Disclaimer: The AI-based summarization tool and virtual assistant provided on this website may not always provide accurate and complete summaries or responses. We encourage you to carefully review and evaluate the generated content to ensure its quality and relevance to your needs.