SkillRAE: Agent Skill-Based Context Compilation for Retrieval-Augmented Execution
2026-05-11 • Computation and Language
Computation and Language
AI summaryⓘ
The authors study how large language model agents use libraries of skills to complete complex tasks. They point out that previous work focused on finding and using skills but not on organizing these skills efficiently for immediate use. To fix this, they created SkillRAE, a method that first builds a detailed skill network offline and then retrieves and compiles the best relevant skills online in a compact and useful way. Their experiments show that SkillRAE improves task performance significantly compared to existing methods, highlighting the importance of how skills are compiled rather than just added as extra prompts.
Large Language ModelsReusable Skill LibrariesRetrieval-Augmented ExecutionSkill GraphContext CompilationOffline IndexingOnline RetrievalSkill-Ranked RetrievalCompact CompilationSkillsBench
Authors
Xiangcheng Meng, Shu Wang, Yixiang Fang
Abstract
Large Language Model (LLM)-based agents (e.g., OpenClaw) increasingly rely on reusable skill libraries to solve artifact-rich tasks such as document-centric workflows and data-intensive analysis. As these libraries grow, a few works have attempted to study the Retrieval-Augmented Execution (RAE), which often first retrieves some external skills and other knowledge, then compiles the context using retrieved skills, and finally executes the task. Existing works mainly focus on optimizing skill retrieval and task execution, and they pay little attention to how to effectively organize the selected skill evidence in a form that is compact, grounded, and immediately usable for the downstream executors to complete tasks. To fill this gap, we propose SkillRAE, a two-stage RAE approach focusing on skill-based context compilation, which consists of the offline and online stages. Specifically, in the offline indexing stage, it builds a multi-level skill graph over skill communities, skills, and reusable subunits, for capturing their relationships. In the online retrieval stage, it first performs skill-ranked retrieval with selected-subunit evidence export in the graph, and then applies rescue-aware compact compilation to recover the key evidence. Together, these components compile a coarse-ranked skill set into a task-specific context that is compact, grounded, and immediately usable. Experiments on two public benchmarks show that SkillRAE achieves a significant improvement over baselines for RAE. For example, on SkillsBench, it achieves an improvement of 11.7% over the SOTA method. Ablation studies further show that our context compilation is crucial, instead of a mere prompt addition.