mlsanityData to 7 Oct 2026

SkillsBench: Benchmarking How Well Agent Skills Work Across Diverse Tasks

arXiv 2602.12670 ↗First posted 13 Feb 2026cs.AIcited by 289

On the map: LLM agents and tool use → Agent skill learning and evolution

Abstract (arXiv)

Agent Skills are structured packages of procedural knowledge that augment large language model (LLM) agents at inference time. Despite rapid adoption, there is no standard way to measure whether they actually help. We present SkillsBench, a benchmark whose current inventory contains 87 tasks across 8 domains paired with curated Skills and deterministic verifiers. Our latest aggregate evaluation runs the 87-task benchmark under matched no-Skills and curated-Skills conditions for 18 model-harness configurations. Curated Skills raise the average pass rate from 33.9% to 50.5% (+16.6 percentage points; 25.5% normalized gain), with configuration-level gains ranging from +4.1 to +25.7 pp. Focused Skills with at most three modules outperform larger or exhaustive bundles, and smaller models with Skills can match larger models without them. SkillsBench establishes paired evaluation as the foundation for rigorous measurement of Skill efficacy on agentic, expertise-heavy work.

Key facts. Cited by 289 papers in the map (data through 7 Oct 2026). Placed in Agent skill learning and evolution, one of 16 directions in LLM agents and tool use. Builds on 38 of its 56 references that are in the map. First posted on arXiv 13 Feb 2026.

Builds on

38 of 56 references.

References · 38

  1. 2020Language Models are Few-Shot Learnerscited by 32,199
  2. 2022Training language models to follow instructions with human feedbackcited by 15,712
  3. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  4. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  5. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,160
  6. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,355
  7. 2022PaLM: Scaling Language Modeling with Pathwayscited by 5,304
  8. 2023Toolformer: Language Models Can Teach Themselves to Use Toolscited by 4,072
  9. 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,064
  10. 2021Program Synthesis with Large Language Modelscited by 3,509
  11. 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 3,317
  12. 2023SWE-bench: Can Language Models Resolve Real-World GitHub Issues?cited by 3,193
26 more
  1. 2023Tree of Thoughts: Deliberate Problem Solving with Large Language Modelscited by 3,155
  2. 2023Voyager: An Open-Ended Embodied Agent with Large Language Modelscited by 2,081
  3. 2022Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language modelscited by 1,970
  4. 2023WebArena: A Realistic Web Environment for Building Autonomous Agentscited by 1,908
  5. 2023ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIscited by 1,880
  6. 2024SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineeringcited by 1,628
  7. 2022Least-to-Most Prompting Enables Complex Reasoning in Large Language Modelscited by 1,397
  8. 2024τ-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domainscited by 1,280
  9. 2023AgentBench: Evaluating LLMs as Agentscited by 1,211
  10. 2024Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preferencecited by 1,151
  11. 2024OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environmentscited by 1,134
  12. 2023DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelinescited by 803
  13. 2023Language Agent Tree Search Unifies Reasoning Acting and Planning in Language Modelscited by 551
  14. 2024BigCodeBench: Benchmarking Code Generation with Diverse Function Calls and Complex Instructionscited by 527
  15. 2026Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line Interfacescited by 515
  16. 2024MLE-bench: Evaluating Machine Learning Agents on Machine Learning Engineeringcited by 429
  17. 2023Cognitive Architectures for Language Agentscited by 422
  18. 2024AppWorld: A Controllable World of Apps and People for Benchmarking Interactive Coding Agentscited by 359
  19. 2025SWE-smith: Scaling Data for Software Engineering Agentscited by 255
  20. 2023InterCode: Standardizing and Benchmarking Interactive Coding with Execution Feedbackcited by 229
  21. 2024Cybench: A Framework for Evaluating Cybersecurity Capabilities and Risks of Language Modelscited by 156
  22. 2022DocPrompting: Generating Code by Retrieving the Docscited by 151
  23. 2019MLPerf Training Benchmarkcited by 114
  24. 2025Establishing Best Practices for Building Rigorous Agentic Benchmarkscited by 94
  25. 2022Execution-Based Evaluation for Open-Domain Code Generationcited by 89
  26. 2025ReplicationBench: Can AI Agents Replicate Astrophysics Research Papers?cited by 10

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 289.

289 papers.

Sort:

By area:

  1. 2026SkillOpt: Executive Strategy for Self-Evolving Agent Skillscited by 122
  2. 2026SoK: Agentic Skills -- Beyond Tool Use in LLM Agentscited by 120
  3. 2026CoEvoSkills: Self-Evolving Agent Skills via Co-Evolutionary Verificationcited by 117
  4. 2026SKILL0: In-Context Agentic Reinforcement Learning for Skill Internalizationcited by 82
  5. 2026SkillClaw: Let Skills Evolve Collectively with Agentic Evolvercited by 76
  6. 2026SWE-Skills-Bench: Do Agent Skills Actually Help in Real-World Software Engineering?cited by 72
  7. 2026SkillX: Automatically Constructing Skill Knowledge Bases for Agentscited by 69
  8. 2026Externalization in LLM Agents: A Unified Review of Memory, Skills, Protocols and Harness Engineeringcited by 68
  9. 2026Organizing, Orchestrating, and Benchmarking Agent Skills at Ecosystem Scalecited by 65
  10. 2026SkillRouter: Skill Routing for LLM Agents at Scalecited by 57
  11. 2026Graph-of-Skills: Dependency-Aware Structural Retrieval for Massive Agent Skillscited by 53
  12. 2026SkillNet: Create, Evaluate, and Connect AI Skillscited by 48
38 more
  1. 2026SkillLearnBench: Benchmarking Continual Learning Methods for Agent Skill Generation on Real-World Taskscited by 45
  2. 2026SkillOS: Learning Skill Curation for Self-Evolving Agentscited by 45
  3. 2026SkillForge: Forging Domain-Specific, Self-Evolving Agent Skills in Cloud Technical Supportcited by 43
  4. 2026Skill Retrieval Augmentation for Agentic AIcited by 41
  5. 2026TCOD: Exploring Temporal Curriculum in On-Policy Distillation for Multi-turn Autonomous Agentscited by 38
  6. 2026SkillRet: A Large-Scale Benchmark for Skill Retrieval in LLM Agentscited by 37
  7. 2026Beyond Semantic Similarity: Rethinking Retrieval for Agentic Search via Direct Corpus Interactioncited by 33
  8. 2026SkillReducer: Optimizing LLM Agent Skills for Token Efficiencycited by 29
  9. 2026Harness Updating Is Not Harness Benefit: Disentangling Evolution Capabilities in Self-Evolving LLM Agentscited by 28
  10. 2026MUSE-Autoskill: Self-Evolving Agents via Skill Creation, Memory, Management, and Evaluationcited by 23
  11. 2026SkillGrad: Optimizing Agent Skills Like Gradient Descentcited by 23
  12. 2026SkillRevise: Improving LLM-Authored Agent Skills via Trace-Conditioned Skill Revisioncited by 21
  13. 2026SkillJect: Effectively Automating Skill-Based Prompt Injection for Skill-Enabled Agentscited by 20
  14. 2026SkillFlow:Benchmarking Lifelong Skill Discovery and Evolution for Autonomous Agentscited by 19
  15. 2026From Context to Skills: Can Language Models Learn from Context Skillfully?cited by 19
  16. 2026SkillsVote: Lifecycle Governance of Agent Skills from Collection, Recommendation to Evolutioncited by 18
  17. 2026From Raw Experience to Skill Consumption: A Systematic Study of Model-Generated Agent Skillscited by 18
  18. 2026A Comprehensive Survey on Agent Skills: Taxonomy, Techniques, and Applicationscited by 17
  19. 2026GEMS: Agent-Native Multimodal Generation with Memory and Skillscited by 16
  20. 2026Experience Compression Spectrum: Unifying Memory, Skills, and Rules in LLM Agentscited by 16
  21. 2026SkillSafetyBench: Evaluating Agent Safety under Skill-Facing Attack Surfacescited by 15
  22. 2026SkVM: Revisiting Language VM for Skills across Heterogenous LLMs and Harnessescited by 14
  23. 2026SkillMOO: Multi-Objective Optimization of Agent Skills for Software Engineeringcited by 14
  24. 2026ContractSkill: Repairable Contract-Based Skills for Multimodal Web Agentscited by 13
  25. 2026Co-Evolving LLM Decision and Skill Bank Agents for Long-Horizon Taskscited by 13
  26. 2026SkillEvolver: Skill Learning as a Meta-Skillcited by 12
  27. 2026MetaSkill-Evolve: Recursive Self-Improvement of LLM Agents via Two-Timescale Meta-Skill Evolutioncited by 12
  28. 2026More Skills, Worse Agents? Skill Shadowing Degrades Performance When Expanding Skill Librariescited by 11
  29. 2026VitaBench 2.0: Evaluating Personalized and Proactive Agents in Long-Term User Interactionscited by 11
  30. 2026SkillAxe: Sharpening LLM-Authored Agent Skills Through Evaluation-Guided Self-Refinementcited by 11
  31. 2026SkillAudit: Ground-Truth-Free Skill Evolution via Paired Trajectory Auditingcited by 11
  32. 2026Library Drift: Diagnosing and Fixing a Silent Failure Mode in Self-Evolving LLM Skill Librariescited by 10
  33. 2026SkillAdaptor: Self-Adapting Skills for LLM Agents from Trajectoriescited by 10
  34. 2026EffiSkill: Agent Skill Based Automated Code Efficiency Optimizationcited by 9
  35. 2026From Procedural Skills to Strategy Genes: Towards Experience-Driven Test-Time Evolutioncited by 9
  36. 2026Group of Skills: Group-Structured Skill Retrieval for Agent Skill Librariescited by 9
  37. 2026SkillMaster: Toward Autonomous Skill Mastery in LLM Agentscited by 8
  38. 2026SkillRAE: Agent Skill-Based Context Compilation for Retrieval-Augmented Executioncited by 8

50 most cited of 289 All 289 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026A Framework for Evaluating Agentic Skills at Scale
  2. 2026How Well Do Agentic Skills Work in the Wild: Benchmarking LLM Skill Usage in Realistic Settings
  3. 2026SkillGenBench: Benchmarking Skill Generation Pipelines for LLM Agents
  4. 2026SkillAudit: From Fixed-Suite Benchmarking to Skill-Centered Assessment
  5. 2026SKT: Skill-Use Training at Scale via Verified Synthetic Data Generation
  6. 2025APTBench: Benchmarking Agentic Potential of Base LLMs During Pre-Training
  7. 2026Agent Skills for Large Language Models: Architecture, Acquisition, Security, and the Path Forward
  8. 2025SkillFlow: Scalable and Efficient Agent Skill Retrieval System