A Systematic Survey of Agentic Skills: Architecture, Lifecycle, and Security
cs.AI, cs.LG, cs.MA
Submitted: 2026-08-30
Updated: 2026-09-28
DOI: 10.2139/ssrn.7245419
License: http://creativecommons.org/licenses/by/4.0/
The gist: Autonomous large language model (LLM) agents increasingly face reliability, context consumption, and execution stability bottlenecks when deployed on complex, long-horizon tasks.
Terminology
Abstract
Autonomous large language model (LLM) agents increasingly face reliability, context consumption, and execution stability bottlenecks when deployed on complex, long-horizon tasks. While monolithic prompt engineering and stateless tool-calling paradigms struggle to scale, the field is rapidly converging toward agentic skills: modular procedural abstractions that externalize execution knowledge into reusable, executable, and portable artifacts. This paper establishes a unified systems foundation and reference architecture for the agentic skills ecosystem. We formalize skills as externalized procedural knowledge bridging high-level cognitive planning with deterministic execution environments, and systematically delineate the architecture across a nine-stage lifecycle: autonomous discovery, authoring and representation formats, memory storage, dynamic retrieval and routing, composition and orchestration, execution and repair, lifelong adaptation, empirical evaluation, and security governance. We further examine marketplace dynamics, public registries, and emerging adversarial threat vectors, alongside runtime verification and defense mechanisms. Finally, we categorize system implementations across software engineering, operating system navigation, embodied robotics, and scientific discovery, while highlighting critical open challenges in continual learning and benchmark realism. This work establishes agentic skills as a foundational paradigm for building scalable, robust, and verifiable autonomous language agents.
Sources
- EvoSkill: Automated Skill Discovery for Multi-Agent Systems
- Agent Skill Security: Threat Models, Attacks, Defenses, and Evaluation
- SKILL.state: Scalable Long-Horizon Agent Skills
- Automating Skill Acquisition through Large-Scale Mining of Open-Source Agentic Repositories: A Framework for Multi-Agent Procedural Knowledge Extraction
- When Skills Don't Help: A Negative Result on Procedural Knowledge for Tool-Grounded Agents in Offensive Cybersecurity
- SkVM: Revisiting Language VM for Skills across Heterogenous LLMs and Harnesses
- SkillDroid: Compile Once, Reuse Forever
- SkillCraft: Can LLM Agents Learn to Use Tools Skillfully?
- Dynamic Malicious Skills in Agentic AI
- CUA-Skill: Develop Skills for Computer Using Agent
- How Your Credentials Are Leaked by LLM Agent Skills: An Empirical Study
- SkillRet: A Large-Scale Benchmark for Skill Retrieval in LLM Agents
- Rethinking Agent Design: From Top-Down Workflows to Bottom-Up Skill Evolution
- Trace2Skill: Verifier-Guided Skill Evolution for Long-Context EDA Agents
- SkillAttack: Automated Red Teaming of Agent Skills through Attack Path Refinement
- Detecting Malicious Agent Skills in the Wild using Attention
- Toward Scalable Terminal Task Synthesis via Skill Graphs
- Memp: Exploring Agent Procedural Memory
- SkillTrojan: Backdoor Attacks on Skill-Based Agent Systems
- SkillReducer: Optimizing LLM Agent Skills for Token Efficiency
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems