自动化构建海量工具数据集，提升大模型使用真实工具的能力。
MCP-Flow: Facilitating LLM Agents to Master Real-World, Diverse and Scaling MCP Tools
- 用网页代理自动发现并筛选1166个服务器的工具数据
- 生成6.8万条高质量指令-调用对和6439条任务轨迹
- 适合想提升大模型实际应用能力的研究者与开发者
大型语言模型日益依赖外部工具完成复杂现实任务，但其在快速扩展的模型上下文协议（MCP）生态中的应用能力仍受限。现有MCP研究覆盖服务器少、依赖高成本人工整理，且缺乏训练支持，阻碍了真实场景部署。为此，我们提出MCP-Flow，一种由网页代理驱动的自动化流水线，实现大规模服务器发现、数据合成与模型训练。MCP-Flow从1166个服务器和11536个工具中收集并过滤数据，生成68733条高质量指令-函数调用对和6439条任务轨迹，规模与多样性远超以往工作。大量实验表明，MCP-Flow显著提升了大模型在工具选择、函数调用生成及智能体任务表现方面的性能。MCP-Flow为大模型智能体在真实MCP环境中的能力提升提供了可扩展的基础。代码已公开于https://github.com/wwh0411/MCP-Flow。
原文摘要 · Abstract (English)
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. MCP-Flow is publicly available at https://github.com/wwh0411/MCP-Flow.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。