An Agent-Oriented Pluggable Experience-RAG Skill for Experience-Driven Retrieval Strategy Orchestration

TL;DR

This paper introduces Experience-RAG Skill, an agent-oriented retrieval orchestration layer that dynamically selects retrieval strategies based on task context, improving retrieval effectiveness across diverse tasks.

Contribution

It proposes a novel pluggable skill for retrieval strategy selection, outperforming fixed retrievers and rivaling adaptive routing methods.

Findings

Achieves nDCG@10 of 0.8924 on multiple datasets.

Outperforms fixed single-retriever baselines.

Remains competitive with adaptive routing approaches.

Abstract

Retrieval-augmented generation systems often assume that one fixed retrieval pipeline is sufficient across heterogeneous tasks, yet factoid question answering, multi-hop reasoning, and scientific verification exhibit different retrieval preferences. We present Experience-RAG Skill, an agent-oriented pluggable retrieval orchestration layer positioned between the agent and the retriever pool. The proposed skill analyzes the current scene, consults an experience memory, selects an appropriate retrieval strategy, and returns structured evidence to the agent. Under a fixed candidate pool, Experience-RAG Skill achieves an overall nDCG@10 of 0.8924 on BeIR/nq, BeIR/hotpotqa, and BeIR/scifact, outperforming fixed single-retriever baselines and remaining competitive with Adaptive-RAG-style routing. The results suggest that retrieval strategy selection can be productively encapsulated as a reusable agent skill rather than being hard-coded in the upper workflow.