Retrieval-Augmented Skill Optimization via Cross-Harness Adaptation

Hugging Face Daily Papers Papers
agent-skills retrieval-augmentation skill-optimization multi-agent benchmark-evaluation

Summary

该论文提出 RASO（检索增强技能优化）框架，通过 Cross-Harness Adaptation 从外部技能语料库中检索先验知识来初始化和迭代优化 agent 技能，在四个 agent 基准和两个模型上显著优于无检索增强的基线方法。

An agent skill is a reusable, actionable natural-language artifact that guides an agent to perform a task effectively under a given harness. Recent studies have explored the optimization of agent skills, contributing to a growing collection of publicly available skills spanning diverse tasks, domains, and harnesses. Despite millions of publicly shared skills, existing skill optimization methods largely overlook this accumulated knowledge, instead relying solely on expensive agent rollouts to iteratively refine skills for a target task. To address this, we propose Retrieval-Augmented Skill Optimization (RASO), a framework that leverages an external skill corpus as prior knowledge throughout skill optimization. RASO retrieves relevant knowledge from existing skills and adapts it to the target task and harness via Cross-Harness Adaptation, accounting for mismatches in both domain and harness. RASO comprises two complementary stages: Retrieval-Augmented Skill Initialization (RASI) constructs a knowledge-grounded initial skill without requiring agent rollouts, while Retrieval-Augmented Skill Update (RASU) iteratively refines the skill by retrieving external knowledge guided by execution feedback. Across four agent benchmarks and two models, extensive experiments show that RASO consistently outperforms baselines without retrieval-augmented skill initialization and updating.
Original Article
View Cached Full Text

Cached at: 10/02/26, 04:28 AM

Paper page - Retrieval-Augmented Skill Optimization via Cross-Harness Adaptation

Source: https://huggingface.co/papers/2609.38024 Authors:

,

,

,

,

,

,

,

,

,

Abstract

Anagentskillisareusable,actionablenatural-languageartifactthatguidesanagenttoperformataskeffectivelyunderagivenharness.Recentstudieshaveexploredtheoptimizationofagentskills,contributingtoagrowingcollectionofpubliclyavailableskillsspanningdiversetasks,domains,andharnesses.Despitemillionsofpubliclysharedskills,existingskilloptimizationmethodslargelyoverlookthisaccumulatedknowledge,insteadrelyingsolelyonexpensiveagentrolloutstoiterativelyrefineskillsforatargettask.Toaddressthis,weproposeRetrieval-AugmentedSkillOptimization(RASO),aframeworkthatleveragesanexternalskillcorpusaspriorknowledgethroughoutskilloptimization.RASOretrievesrelevantknowledgefromexistingskillsandadaptsittothetargettaskandharnessviaCross-HarnessAdaptation,accountingformismatchesinbothdomainandharness.RASOcomprisestwocomplementarystages:Retrieval-AugmentedSkillInitialization(RASI)constructsaknowledge-groundedinitialskillwithoutrequiringagentrollouts,whileRetrieval-AugmentedSkillUpdate(RASU)iterativelyrefinestheskillbyretrievingexternalknowledgeguidedbyexecutionfeedback.Acrossfouragentbenchmarksandtwomodels,extensiveexperimentsshowthatRASOconsistentlyoutperformsbaselineswithoutretrieval-augmentedskillinitializationandupdating.

View arXiv pageView PDFAdd to collection

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2609.38024 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2609.38024 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2609.38024 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

Google Harness² for Recursive Self Improvement (GitHub Repo)

TLDR AI

Google Research 开源 Harness²，一个实现递归式智能体系统优化（recursive agent harnessing）的即插即用框架，可在模型权重固定的情况下通过 Propose–Probe–Compose 循环持续改进 agent harness，在多个专业基准上带来最高 11.8 个百分点的性能提升。

Training the Whole Agent, Not Just the Model

Hugging Face Daily Papers

HarnessAdapt-RL jointly trains a coding policy and adapts its evaluation harness via mechanism-level skills and evolving tools, improving pass@1 over GRPO by up to 12.6 points on SWE-bench variants with a 9B base model.