# Skill-to-LoRA: From Using Skills to Learning Behaviors for Token-Efficient LLM Agents

arXiv: https://arxiv.org/abs/2606.16769 · HTML: https://mlsanity.com/paper/2606.16769/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Placed in Agent skill learning and evolution, one of 16 directions in LLM agents and tool use.
- Builds on 12 of its 25 references that are in the map.
- First posted on arXiv 15 Jun 2026.

## Placed in

- LLM agents and tool use / [Agent skill learning and evolution](https://mlsanity.com/agents/agent-skill-learning-and-evolution/)

## Builds on (most cited first, 12 of 12)

- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023)
- [FitNets: Hints for Thin Deep Nets](https://mlsanity.com/paper/1412.6550/) (2014)
- [SkillRL: Evolving Agents via Recursive Skill-Augmented Reinforcement Learning](https://mlsanity.com/paper/2602.08234/) (2026)
- [SKILL0: In-Context Agentic Reinforcement Learning for Skill Internalization](https://mlsanity.com/paper/2604.02268/) (2026)
- [SWE-Skills-Bench: Do Agent Skills Actually Help in Real-World Software Engineering?](https://mlsanity.com/paper/2603.15401/) (2026)
- [Text-to-LoRA: Instant Transformer Adaption](https://mlsanity.com/paper/2506.06105/) (2025)
- [SkillLearnBench: Benchmarking Continual Learning Methods for Agent Skill Generation on Real-World Tasks](https://mlsanity.com/paper/2604.20087/) (2026)
- [Doc-to-LoRA: Learning to Instantly Internalize Contexts](https://mlsanity.com/paper/2602.15902/) (2026)
- [Skill Set Optimization: Reinforcing Language Model Behavior via Transferable Skills](https://mlsanity.com/paper/2402.03244/) (2024)
- [Training Plug-n-Play Knowledge Modules with Deep Context Distillation](https://mlsanity.com/paper/2503.08727/) (2025)

## Built on by (most cited first, 10 of 10)

- [SkillZip: Evaluation-Free Skill Compression for Self-Evolving Agents by Discovering Reusable Structure](https://mlsanity.com/paper/2608.11079/) (2026)
- [SKT: Skill-Use Training at Scale via Verified Synthetic Data Generation](https://mlsanity.com/paper/2608.02287/) (2026)
- [Graph Engineering in the Era of LLM Agents: From Individual Intelligence to System Intelligence](https://mlsanity.com/paper/2608.21156/) (2026)
- [From Execution to Capability: Scientific Experience Consolidation via Procedural Knowledge Synthesis](https://mlsanity.com/paper/2607.24459/) (2026)
- [AUSO: Action-Level Unified Skill Optimization from Internalization to Utilization](https://mlsanity.com/paper/2608.21292/) (2026)
- [SkillZip Pro: Execution-Aware Dynamic Compression of Progressively Loaded Skills for Self-Evolving Agents](https://mlsanity.com/paper/2608.30785/) (2026)
- [Test-Driven Approaches to Software Engineering with Large Language Models: A Survey of Phases, Tasks, and Agent Skills](https://mlsanity.com/paper/2609.12012/) (2026)
- [SkillGym: Training Skill-Use Agents with Automatic Verifiable Environment Generation](https://mlsanity.com/paper/2609.37539/) (2026)
- [SkillFM: Generating Skills for LLM Agents via Latent Flow Matching](https://mlsanity.com/paper/2609.39382/) (2026)
- [When Skills Remain But Access Fails: Factorized Skill Memory for Continual Skill Internalization](https://mlsanity.com/paper/2610.15492/) (2026)

## Abstract (arXiv)

Agent skills are commonly distributed as SKILL.md files: human-readable procedural documents that describe workflows, tools, resources, and domain conventions. While convenient for inspection and reuse, this design requires the same reusable procedure to be repeatedly injected into the runtime context. We propose Skill-to-LoRA(S2L), a behavior-centric skill representation that replaces runtime skill text with skill-specific LoRA adapters. Rather than compressing the skill document itself, S2L models the behavioral change induced by the skill text: offline, the complete SKILL.md is used to synthesize skill-guided demonstrations; online, the full document is omitted and the corresponding LoRA adapter is dynamically loaded to activate the learned skill behavior. We evaluate S2L with Qwen3.6-27B on a 21-skill subset of SWE-Skills-Bench. Compared with the no-skill and Full Skill Text baselines, S2L improves pass rate by 2.9 and 5.2 percentage points, respectively, while reducing per-step token cost by 6.6% relative to Full Skill Text prompting. S2L matches or improves Full Skill Text on 18/21 skills and the no-skill baseline on 15/21 skills. Control experiments further show that the gains depend on skill-specific adapter alignment: Wrong-LoRA and Shared-LoRA both reduce performance. These results suggest that many procedural agent skills can be converted from runtime instructions into trainable, dynamically loadable behavioral modules. Code will be released upon acceptance.