Is this server read‑only?
Yes. The README explicitly states it is a read‑only MCP server. For full read, write, update, and delete capabilities, use the separate CData MCP Server for Google Data Catalog (beta) available on the CData website.
@CDataSoftware
基本信息
This project builds a read-only MCP server. For full read, write, update, delete, and action capabilities and a simplified setup, check out our free CData MCP Server for Google Data Catalog (beta): https://www.cdata.com/download/download.aspx?sku=HGZK-V&type=beta
使用下面的配置,将此服务器添加到你的 MCP 客户端。
{
"mcpServers": {
"{classname_dash}": {
"command": "PATH\\TO\\java.exe",
"args": [
"-jar",
"PATH\\TO\\CDataMCP-jar-with-dependencies.jar",
"PATH\\TO\\google-data-catalog.prp"
]
}
}
}工具列表暂未提供
当前收录暂时没有可展示的工具列表,不代表服务没有工具。请查阅项目文档了解支持的能力。
工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。
A read-only Model Context Protocol (MCP) server that enables large language models (LLMs) to query live data from Google Data Catalog. It is built on the CData JDBC Driver for Google Data Catalog, which exposes the data source as relational SQL models. The server wraps that driver in an MCP interface, allowing LLMs to retrieve information using natural language without writing SQL.
Build the server with Maven (mvn clean install), download and install the CData JDBC Driver for Google Data Catalog, license the driver, and create a .prp configuration file with your JDBC connection string. Then add the server to your Claude Desktop configuration (claude_desktop_config.json) pointing to the built JAR and the .prp file. Run or refresh Claude Desktop to use the tools.
get_tables, get_columns, run_query)stdio transport (client must be on same machine)Yes. The README explicitly states it is a read‑only MCP server. For full read, write, update, and delete capabilities, use the separate CData MCP Server for Google Data Catalog (beta) available on the CData website.
You need Java to run the JAR file and Maven to build the project. Additionally, the CData JDBC Driver for Google Data Catalog must be downloaded, installed, and licensed separately.
Connections are defined in a `.prp` file that includes the JDBC URL, driver path, driver class, and optionally a list of tables to expose. The JDBC connection string can be generated using the driver’s connection string utility, which supports OAuth-based authentication.
The data stays in your Google Data Catalog instance. The server never stores or copies the data; it queries it live through the JDBC driver.
It uses `stdio` transport, so the MCP client must run on the same machine as the server. This means remote clients cannot connect directly to this server.

Session replay, built for agents. Subtext is agentic session review: it captures production sessions of your app and connects them to your coding agent — Claude Code, Cursor, Codex, Devin, your own harness — so it can

Server-side conversion tracking for Meta, TikTok, etc. Set up tracking and ask about events.

Validated aviation reference tools for AI agents: resolve airports, airlines, aircraft types, registrations and flight designators to canonical entities with confidence, alternates, validity dates and per-field provenanc

Ask Google Search Console in plain language from Claude, ChatGPT, Cursor or Claude Code. Connect your Google account once and AskWatch issues a personal link: no Cloud project, no credentials.json, nothing to run or host

Building permit data and Florida contractor license checks over MCP. 81.8M permit records from the cities and counties that issue them, across 51 states. Four tools: verify_license, search_permits, lookup_contractor, roo
评论