Repairing Tool Calls Using Post-tool Execution Reflection and RAG

TL;DR

This paper introduces a reflection-based approach combining LLMs and RAG to automatically repair tool calls in agentic systems, significantly improving success rates and query accuracy, especially with troubleshooting documents.

Contribution

We develop a novel post-tool execution reflection method that leverages LLMs and RAG to repair and improve tool calls in agentic systems, focusing on kubectl commands.

Findings

55% pass rate improvement for tool call success

36% increase in correct query answering

Troubleshooting docs outperform official documentation by 10%

Abstract

Agentic systems interact with external systems by calling tools such as Python functions, REST API endpoints, or command line tools such as kubectl in Kubernetes. These tool calls often fail for various syntactic and semantic reasons. Some less obvious semantic errors can only be identified and resolved after analyzing the tool's response. To repair these errors, we develop a post-tool execution reflection component that combines large language model (LLM)-based reflection with domain-specific retrieval-augmented generation (RAG) using documents describing both the specific tool being called and troubleshooting documents related to the tool. For this paper, we focus on the use case of the kubectl command line tool to manage Kubernetes, a platform for orchestrating cluster applications. Through a larger empirical study and a smaller manual evaluation, we find that our RAG-based reflection will repair kubectl commands such that they are both more likely to successfully execute (pass rate) for 55% of our models evaluated and 36% more likely to correctly answer the user query on average. We find that troubleshooting documents improve pass rate compared to official documentation by an average of 10%.