MCP-Flow: Facilitating LLM Agents to Master Real-World, Diverse and Scaling MCP Tools

WenHao Wang, Peizhi Niu, Zhao Xu, Zhaoyu Chen, Jian Du, Yaxin Du, Xianghe Pang, Keduan Huang, Yanfeng Wang, Qiang Yan, Siheng Chen

Keywords: Model Contextual Protocol, Large Language Model, Tool Learning, Funcation Calling, Dataset Construction

TL;DR: We propose MCP-Flow, a comprehensive pipeline, data and model suite that automatically constructs datasets from real-world, diverse and continuously scaling MCP servers, thereby facilitating more effective utilization of MCP tools by LLM agents.

Abstract:

Large Language Models (LLMs) increasingly rely on external tools to perform complex, realistic tasks, yet their ability to utilize the rapidly expanding Model Contextual Protocol (MCP) ecosystem remains limited. Existing MCP research covers few servers, depends on costly manual curation, and lacks training support, hindering progress toward real-world deployment.
To overcome these limitations, we introduce MCP-Flow, an automated web-agent-driven pipeline for large-scale server discovery, data synthesis, and model training.
MCP-Flow collects and filters data from 1166 servers and 11536 tools, producing 68733 high-quality instruction-function call pairs and 6439 trajectories, far exceeding prior work in scale and diversity.
Extensive experiments demonstrate MCP-Flow's effectiveness in driving superior MCP tool selection, function-call generation, and enhanced agentic task performance.
MCP-Flow thus provides a scalable foundation for advancing LLM agents' proficiency in real-world MCP environments.

Primary Area: applications to computer vision, audio, language, and other modalities

Submission Number: 9547