Data to 7 Oct 2026

Inducing Programmatic Skills for Agentic Tasks

arXiv 2504.06821 ↗First posted 9 Apr 2025cs.CLcited by 103

On the map: LLM agents and tool use → Web, search, and research agents

Abstract (arXiv)

To succeed in common digital tasks such as web navigation, agents must carry out a variety of specialized tasks such as searching for products or planning a travel route. To tackle these tasks, agents can bootstrap themselves by learning task-specific skills online through interaction with the web environment. In this work, we demonstrate that programs are an effective representation for skills. We propose agent skill induction (ASI), which allows agents to adapt themselves by inducing, verifying, and utilizing program-based skills on the fly. We start with an evaluation on the WebArena agent benchmark and show that ASI outperforms the static baseline agent and its text-skill counterpart by 23.5% and 11.3% in success rate, mainly thanks to the programmatic verification guarantee during the induction phase. ASI also improves efficiency by reducing 10.7-15.3% of the steps over baselines, by composing primitive actions (e.g., click) into higher-level skills (e.g., search product). We then highlight the efficacy of ASI in remaining efficient and accurate under scaled-up web activities. Finally, we examine the generalizability of induced skills when transferring between websites, and find that ASI can effectively reuse common skills, while also updating incompatible skills to versatile website changes.

Key facts. Cited by 103 papers in the map (data through 7 Oct 2026). Placed in Web, search, and research agents, one of 16 directions in LLM agents and tool use. Builds on 41 of its 46 references that are in the map. First posted on arXiv 9 Apr 2025.

Builds on

41 of 46 references.

Closest ancestors · 13

Same direction, newest first

1 more

2022WebShop: Towards Scalable Real-World Web Interaction with Grounded Language Agents4 Jul

Other references · 28

2020Language Models are Few-Shot Learnerscited by 32,199
2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,064
2023Voyager: An Open-Ended Embodied Agent with Large Language Modelscited by 2,081
2020Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problemscited by 1,873
2024τ-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domainscited by 1,280
2022Code as Policies: Language Model Programs for Embodied Controlcited by 1,172
2024OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environmentscited by 1,134
2024TravelPlanner: A Benchmark for Real-World Planning with Language Agentscited by 385
2023Language to Rewards for Robotic Skill Synthesiscited by 307
2024TheAgentCompany: Benchmarking LLM Agents on Consequential Real World Taskscited by 306
2023Large Language Models as Tool Makerscited by 293
2018Reinforcement Learning on Web Interfaces Using Workflow-Guided Explorationcited by 282

16 more

2024Agent Q: Advanced Reasoning and Learning for Autonomous AI Agentscited by 201
2024Trial and Error: Exploration-Based Trajectory Optimization for LLM Agentscited by 194
2023Synapse: Trajectory-as-Exemplar Prompting with Memory for Computer Controlcited by 180
2024Recursive Introspection: Teaching Language Model Agents How to Self-Improvecited by 162
2020DreamCoder: Growing generalizable, interpretable knowledge with wake-sleep Bayesian program learningcited by 159
2024OmniACT: A Dataset and Benchmark for Enabling Multimodal Generalist Autonomous Agents for Desktop and Webcited by 156
2021Skill Induction and Planning with Latent Languagecited by 99
2023CLIN: A Continually Learning Language Agent for Rapid Task Adaptation and Generalizationcited by 88
2025Scaling Test-Time Compute Without Verification or RL is Suboptimalcited by 82
2024TroVE: Inducing Verifiable and Efficient Toolboxes for Solving Programmatic Taskscited by 70
2018Learning to Navigate the Webcited by 60
2024VLM Agents Generate Their Own Memories: Distilling Experience into Embodied Programs of Thoughtcited by 55
2023LILO: Learning Interpretable Libraries by Compressing and Documenting Codecited by 46
2024Synatra: Turning Indirect Knowledge into Direct Demonstrations for Digital Agents at Scalecited by 43
2019Program Synthesis and Semantic Parsing with Learned Code Idiomscited by 34
2024BAGEL: Bootstrapping Agents by Guiding Exploration with Languagecited by 34

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 21; 2026: 82.

103 papers.

Sort:

By area:

2025Memory in the Age of AI Agentscited by 285
2025ReasoningBank: Scaling Agent Self-Evolving with Reasoning Memorycited by 253
2025Reinforcement Learning for Self-Improving Agent with Skill Librarycited by 91
2026Agentic Memory: Learning Unified Long-Term and Short-Term Memory Management for Large Language Model Agentscited by 72
2026XSkill: Continual Learning from Experience and Skills in Multimodal Agentscited by 71
2026SkillX: Automatically Constructing Skill Knowledge Bases for Agentscited by 69
2026Externalization in LLM Agents: A Unified Review of Memory, Skills, Protocols and Harness Engineeringcited by 68
2026How Well Do Agentic Skills Work in the Wild: Benchmarking LLM Skill Usage in Realistic Settingscited by 62
2026Graph-of-Skills: Dependency-Aware Structural Retrieval for Massive Agent Skillscited by 53
2026SkillLearnBench: Benchmarking Continual Learning Methods for Agent Skill Generation on Real-World Taskscited by 45
2026SkillOS: Learning Skill Curation for Self-Evolving Agentscited by 45
2025DeepShop: A Benchmark for Deep Research Shopping Agentscited by 42

38 more

2026A Survey of Agent Memory in the Second Half: Towards Self-Evolving and Long-Horizon Agentscited by 40
2025Thinking vs. Doing: Agents that Reason by Scaling Test-Time Interactioncited by 34
2025Go-Browse: Training Web Agents with Structured Explorationcited by 32
2025PolySkill: Learning Generalizable Skills Through Polymorphic Abstractioncited by 25
2025Accumulating Context Changes the Beliefs of Language Modelscited by 25
2026GraSP: Graph-Structured Skill Compositions for LLM Agentscited by 25
2026WebXSkill: Skill Learning for Autonomous Web Agentscited by 24
2026SkillTrojan: Backdoor Attacks on Skill-Based Agent Systemscited by 21
2025Adaptation of Agentic AI: A Survey of Post-Training, Memory, and Skillscited by 19
2026From Storage to Experience: A Survey on the Evolution of LLM Agent Memory Mechanismscited by 19
2026A Comprehensive Survey on Agent Skills: Taxonomy, Techniques, and Applicationscited by 17
2025WALT: Web Agents that Learn Toolscited by 15
2025WebCoach: Self-Evolving Web Agents with Cross-Session Memory Guidancecited by 15
2026SkillSafetyBench: Evaluating Agent Safety under Skill-Facing Attack Surfacescited by 15
2026ContractSkill: Repairable Contract-Based Skills for Multimodal Web Agentscited by 13
2026Skill-R1: Agent Skill Evolution via Reinforcement Learningcited by 12
2025Sample-Efficient Online Learning in LM Agents via Hindsight Trajectory Rewritingcited by 10
2025R-WoM: Retrieval-augmented World Model For Computer-use Agentscited by 10
2025AgentProg: Empowering Long-Horizon GUI Agents with Program-Guided Context Managementcited by 10
2026How Well Does Agent Development Reflect Real-World Work?cited by 10
2025MAS-Bench: A Unified Benchmark for Shortcut-Augmented Hybrid Mobile GUI Agentscited by 8
2026Combee: Scaling Prompt Learning for Self-Improving Language Model Agentscited by 8
2026SkillGraph: Self-Evolving Multi-Agent Collaboration with Multimodal Graph Topologycited by 8
2026Generate, Filter, Control, Replay: A Comprehensive Survey of Rollout Strategies for LLM Reinforcement Learningcited by 8
2026Agentic Environment Engineering for Large Language Models: A Survey of Environment Modeling, Synthesis, Evaluation, and Applicationcited by 8
2026Evolving Programmatic Skill Networkscited by 7
2026Lifting Traces to Logic: Programmatic Skill Induction with Neuro-Symbolic Learning for Long-Horizon Agentic Taskscited by 7
2026Better Harnesses, Smaller Models: Building 90% Cheaper Agents via Automated Harness Adaptationcited by 7
2026Are Online Skill and Memory Modules Always Worth Their Tokens? A Budget-Constrained Study of Web Agentscited by 6
2026Dynamic Agent Skills: A Lifecycle Survey and Taxonomy of Evolving Skill Librariescited by 6
2025ToolLibGen: Scalable Automatic Tool Creation and Aggregation for LLM Reasoningcited by 5
2026Chain-of-Memory: Lightweight Memory Construction with Dynamic Evolution for LLM Agentscited by 5
2026Co-Evolving Skill Generation and Policy Optimizationcited by 5
2026TARSE: Test-Time Adaptation via Retrieval of Skills and Experience for Reasoning Agentscited by 4
2026The Tool Illusion: Rethinking Tool Use in Web Agentscited by 3
2026VISUALSKILL: Multimodal Skills for Computer-Use Agentscited by 3
2026Humans are Missing from AI Coding Agent Researchcited by 3
2025RefTool: Reference-Guided Tool Creation for Knowledge-Intensive Reasoningcited by 2

50 most cited of 103 All 103 (JSON)

Similar, not cited

Close in wording, no citation either way