Agent-ValueBench is the first dedicated benchmark for agent values, showing they diverge from LLM values, form a homogeneous 'Value Tide' across models, and bend under harnesses and skill steering.
hub Mixed citations
Natural-Language Agent Harnesses
Mixed citation behavior. Most common role is background (67%).
27
Pith papers citing it
1
external citations · Pith
Background
67% of classified citations
abstract
Agent performance is strongly shaped by the surrounding harness: the external execution system around a model that organizes a task run. Yet this logic is usually buried in tightly coupled controller code, which makes harnesses hard to inspect, compare, transfer, and ablate. This paper asks whether the reusable design pattern of an agent harness can be represented as an executable natural-language object. We introduce Natural-Language Agent Harnesses (NLAHs), editable documents that describe run-level harness policy, and Intelligent Harness Runtime (IHR), a shared runtime that interprets these documents into agent calls, handoffs, state updates, validation gates, and artifact contracts. Across coding, terminal-use, and computer-use benchmarks, IHR-executed NLAHs achieve comparable task outcomes to code and prompted realizations, while exposing much shorter static harness policies. Module ablations further show that explicit harness modules are analyzable. These results suggest that agent harnesses can be turned from incidental glue around models into scientific representation objects.
hub tools
citation-role summary
background 6
baseline 2
method 1
citation-polarity summary
years
2026 27representative citing papers
OCELOT recasts agent privacy as posterior-risk control and implements Witness-Verified Declassification to authorize the least-disclosing useful release under a sink-trust-weighted min-entropy budget.
A self-evolving multi-agent catalysis digital twin (CatDT) matches seven gas–solid kinetic benchmarks within ~0.5–2× experiment and computationally proposes non-precious PDH candidates including Ni@ZrO2.
SkillHarm benchmark shows current AI agents are vulnerable to lifecycle-aware skill poisoning with success rates up to 86.3% for fixed-payload attacks and 69.3% for self-mutating attacks.
Memory for long-horizon agents should preserve distinctions that affect decisions under a fixed budget, not descriptive features, yielding an exact forgetting boundary and a new online learner DeMem with regret guarantees.
LLM agents in a solver-aware harness recover global constraints from MIP formulations, generate executable propagation-only handlers for SCIP, and solve five additional MIPLIB 2017 instances.
OPHSD uses harness-augmented models as teachers to distill reasoning capabilities into base LLMs, yielding strong standalone performance on classification and math tasks.
COMFYCLAW introduces skill evolution via graph editing, automatic reversion, VLM verification, and distillation of runs into reusable Agent Skills, achieving higher average scores than a verifier-only baseline across benchmarks.
An agent harness combining staged task decomposition, multimodal evidence tooling, and artifact-grounded self-improvement scores 81.0 GRAS on multimodal scientific curation, 22.4 points above the strongest baseline — with the caveat that 8 of 23 evaluation papers were used for optimization.
BioInsight is a multi-agent orchestration system that converts disease-associated protein tables into interactive evidence dashboards with typed artifact contracts, improving reported traceability and ranking over baselines given the same retrieved evidence.
SEAGym turns existing benchmarks into multi-view evaluation sources for measuring reusable improvements in LLM agent harnesses, revealing complementary signals missed by single-curve or isolated-task tests.
AURA improves implicit-need coverage by 0.07 over ReAct baselines on a 100-query benchmark by inserting an intent inference step controlled by a gap score, while cutting probes 82% on factual tasks.
A dual-plane permission framework for LLM agent skills cuts SkillInject attack success by roughly 9 percentage points while nearly preserving benign task success.
Catalogs ten patterns and synthesizes a four-layer reference architecture for skill harnessing in LLM agents, evaluated via cross-instantiation on eight systems.
A new LLM-based system generates and validates runnable modular MCMC samplers directly from natural-language Bayesian model descriptions, reporting success on 120 of 132 benchmark models.
Partial harnesses for LLM agents, specifying only initial execution steps, achieve higher pass rates than fully decomposed workflows, as analyzed through trajectory alignment and validated in synthetic and terminal benchmarks.
A typed-contract harness with containerized 'chambers' and robotics-specific agent skills lets a coding LLM turn a single natural-language prompt into working reproduction, evaluation, and deployment workflows for robot learning.
AutoPyVerifier learns compact sets of executable Python verifiers from labeled LLM outputs via LLM synthesis and DAG search, improving objective prediction by up to 55 F1 points and downstream LLM accuracy by up to 17 points.
Agent-World autonomously synthesizes verifiable real-world tasks and uses continuous self-evolution to train 8B and 14B agents that outperform proprietary models on 23 benchmarks.
Robot middleware is the harness for Physical AI and should implement Projection, Isolation, and Transfer to enforce AI model outputs across control, computation, and communication.
AION is a time series harness using agents, skills, rules, memory, evaluation, and protocols with temporal grounding, shown in a Kaggle Store Sales case study to produce more artifacts and reviews than direct agent use.
A survey that organizes existing work on LLM-based agents around code as the central harness, structured in three layers of interfaces, mechanisms, and multi-agent scaling, with applications across domains and listed open challenges.
Categorical Architecture triple (G, Know, Phi) supplies the formal theory for composing LLM agent harnesses with structurally preserved certificates.
A systems-level data model for preserving typed, addressable, versioned, and dependency-aware intermediate artifacts in agentic AI systems to improve long-term inspectability and maintainability.
citing papers explorer
-
Agent-ValueBench: A Comprehensive Benchmark for Evaluating Agent Values
Agent-ValueBench is the first dedicated benchmark for agent values, showing they diverge from LLM values, form a homogeneous 'Value Tide' across models, and bend under harnesses and skill steering.
-
OCELOT: Inference-Leakage Budgets for Privacy-Preserving LLM Agents
OCELOT recasts agent privacy as posterior-risk control and implements Witness-Verified Declassification to authorize the least-disclosing useful release under a sink-trust-weighted min-entropy budget.
-
Autonomous heterogeneous catalyst discovery with a self-evolving multi-agent digital twin
A self-evolving multi-agent catalysis digital twin (CatDT) matches seven gas–solid kinetic benchmarks within ~0.5–2× experiment and computationally proposes non-precious PDH candidates including Ni@ZrO2.
-
SkillHarm: Lifecycle-Aware Skill-Based Attacks via Automated Construction
SkillHarm benchmark shows current AI agents are vulnerable to lifecycle-aware skill poisoning with success rates up to 86.3% for fixed-payload attacks and 69.3% for self-mutating attacks.
-
Remember the Decision, Not the Description: A Rate-Distortion Framework for Agent Memory
Memory for long-horizon agents should preserve distinctions that affect decisions under a fixed budget, not descriptive features, yielding an exact forgetting boundary and a new online learner DeMem with regret guarantees.
-
Agentic MIP Research: Accelerated Constraint Handler Generation
LLM agents in a solver-aware harness recover global constraints from MIP formulations, generate executable propagation-only handlers for SCIP, and solve five additional MIPLIB 2017 instances.
-
Training with Harnesses: On-Policy Harness Self-Distillation for Complex Reasoning
OPHSD uses harness-augmented models as teachers to distill reasoning capabilities into base LLMs, yielding strong standalone performance on classification and math tasks.
-
COMFYCLAW: Self-Evolving Skill Harnesses for Image Generation Workflows
COMFYCLAW introduces skill evolution via graph editing, automatic reversion, VLM verification, and distillation of runs into reusable Agent Skills, achieving higher average scores than a verifier-only baseline across benchmarks.
-
Building Agent Harnesses for Scientific Curation from Multimodal Sources
An agent harness combining staged task decomposition, multimodal evidence tooling, and artifact-grounded self-improvement scores 81.0 GRAS on multimodal scientific curation, 22.4 points above the strongest baseline — with the caveat that 8 of 23 evaluation papers were used for optimization.
-
BioInsight: Multi-Agent Orchestration for Interactive Biomedical Knowledge Discovery
BioInsight is a multi-agent orchestration system that converts disease-associated protein tables into interactive evidence dashboards with typed artifact contracts, improving reported traceability and ranking over baselines given the same retrieved evidence.
-
SEAGym: An Evaluation Environment for Self-Evolving LLM Agents
SEAGym turns existing benchmarks into multi-view evaluation sources for measuring reusable improvements in LLM agent harnesses, revealing complementary signals missed by single-curve or isolated-task tests.
-
AURA: Intent-Directed Probing for Implicit-Need Surfacing in Situated LLM Agents
AURA improves implicit-need coverage by 0.07 over ReAct baselines on a 100-query benchmark by inserting an intent inference step controlled by a gap score, while cutting probes 82% on factual tasks.
-
SkillGuard: A Permission-Centric Framework for Agent Skill Security
A dual-plane permission framework for LLM agent skills cuts SkillInject attack success by roughly 9 percentage points while nearly preserving benign task success.
-
Harnessing Agent Skills: Architectural Patterns and a Reference Architecture for Skill-Mediated LLM Agents
Catalogs ten patterns and synthesizes a four-layer reference architecture for skill harnessing in LLM agents, evaluated via cross-instantiation on eight systems.
-
AI4BayesCode: From Natural Language Descriptions to Validated Modular Stateful Bayesian Samplers
A new LLM-based system generates and validates runnable modular MCMC samplers directly from natural-language Bayesian model descriptions, reporting success on 120 of 132 benchmark models.
-
Harnesses for Inference-Time Alignment over Execution Trajectories
Partial harnesses for LLM agents, specifying only initial execution steps, achieve higher pass rates than fully decomposed workflows, as analyzed through trajectory alignment and validated in synthetic and terminal benchmarks.
-
Nautilus: From One Prompt to Plug-and-Play Robot Learning
A typed-contract harness with containerized 'chambers' and robotics-specific agent skills lets a coding LLM turn a single natural-language prompt into working reproduction, evaluation, and deployment workflows for robot learning.
-
AutoPyVerifier: Learning Compact Executable Verifiers for Large Language Model Outputs
AutoPyVerifier learns compact sets of executable Python verifiers from labeled LLM outputs via LLM synthesis and DAG search, improving objective prediction by up to 55 F1 points and downstream LLM accuracy by up to 17 points.
-
Agent-World: Scaling Real-World Environment Synthesis for Evolving General Agent Intelligence
Agent-World autonomously synthesizes verifiable real-world tasks and uses continuous self-evolution to train 8B and 14B agents that outperform proprietary models on 23 benchmarks.
-
Harness Engineering for Physical AI: Robot Middleware Is the Harness Layer
Robot middleware is the harness for Physical AI and should implement Projection, Isolation, and Transfer to enforce AI model outputs across control, computation, and communication.
-
AION: Next-Generation Tasks and Practical Harness for Time Series
AION is a time series harness using agents, skills, rules, memory, evaluation, and protocols with temporal grounding, shown in a Kaggle Store Sales case study to produce more artifacts and reviews than direct agent use.
-
Code as Agent Harness
A survey that organizes existing work on LLM-based agents around code as the central harness, structured in three layers of interfaces, mechanisms, and multi-agent scaling, with applications across domains and listed open challenges.
-
Harness Engineering as Categorical Architecture
Categorical Architecture triple (G, Know, Phi) supplies the formal theory for composing LLM agent harnesses with structurally preserved certificates.
-
Intermediate Artifacts as First-Class Citizens: A Data Model for Durable Intermediate Artifacts in Agentic Systems
A systems-level data model for preserving typed, addressable, versioned, and dependency-aware intermediate artifacts in agentic AI systems to improve long-term inspectability and maintainability.
-
From Agent Loops to Deterministic Graphs: Execution Lineage for Reproducible AI-Native Work
Execution lineage models AI-native work as a DAG of computations with explicit dependencies, achieving perfect state preservation in controlled update tasks where loop-based agents introduce churn and contamination.
-
From Question Answering to Task Completion: A Survey on Agent System and Harness Design
Survey framing LLM agents as model-plus-harness systems, decomposing harness responsibilities, mapping them to tasks, and highlighting open challenges in evaluation, safety, and co-evolution.
-
Agentic World Modeling: Foundations, Capabilities, Laws, and Beyond
A survey proposing a three-level capability taxonomy (L1 Predictor, L2 Simulator, L3 Evolver) for world models across physical, digital, social, and scientific domains.