MCP 服务器
开放的 Model Context Protocol 服务器注册表。按类别、传输方式或使用场景,为 AI 智能体寻找合适的工具、资源和提示词。
服务器
177
工具
471
类别
11
贡献者
155
Provides AI agents with access to ClickHouse analytical databases through the Model Context Protocol. Enables running analytical queries, exploring table schemas, inspecting materialized views, and monitoring query performance. Designed for OLAP workloads with support for large result sets and query profiling.
MCP server for Google BigQuery that lets AI agents explore datasets, inspect table schemas, and run SQL analytics queries with dry-run cost estimation. Useful for natural-language data analysis, ad-hoc reporting, and pipeline debugging against large-scale BigQuery warehouses.
Official Snowflake MCP server enabling AI agents to query and analyze data in the Snowflake AI Data Cloud. Supports running SQL against warehouses, exploring databases and schemas, describing tables, and invoking Cortex AI services for search and analytics, with role-based access control honored end to end.
Apache Doris' MCP server provides agents with governed access to Doris analytics databases for schema discovery, query assistance, and operational investigation.
Skills 与 MCP 服务器
有什么区别?Skills定义“做什么”
Skill 将说明、示例提示词和推荐模型组织在一起,让智能体稳定地完成任务。Skills 增加的是知识,而不是新的连接。
MCP 服务器定义“如何连接”
MCP 服务器通过连接数据库、API 和文件等真实系统,为智能体增加新能力。MCP 提供连接和操作,而不是任务说明。
简单来说:当模型需要把某项任务做好时选择 Skill;当模型需要访问工具或系统时选择 MCP 服务器。两者可以组合使用,Skill 可以依赖 MCP 服务器提供的工具。
开发了 MCP 服务器?
将它提交到由社区维护的开源注册表。