greenhouse_get_tablesLABSCO SUMMARY
Three tools, any table in Greenhouse.
The whole point is the shape of the tool surface: three generic tools instead of one tool per Greenhouse object, so the context cost stays flat no matter how wide the schema is and the model discovers the schema by asking. The price is setup — a Maven build plus a licensed JDBC driver before anything runs — and read-only access. CData points anyone who wants writes, actions or a hosted endpoint at its commercial products.
What it is
A local, read-only MCP server that CData builds on top of its JDBC driver for Greenhouse. The driver presents Greenhouse as relational SQL models; this server wraps the driver and hands an assistant a small SQL interface over it, so questions get answered from live data instead of a stale export. It runs as a Java process over stdio on your own machine.
What you get
- Three tools, prefixed with the server name you set in the properties file: one lists the tables the driver exposes, one lists a table's columns, and one runs a SQL SELECT.
- Table and column output comes back as CSV with a header row, so the assistant can orient itself on an unfamiliar schema before it queries anything.
- Read access only — the query tool takes SELECT. Nothing in this server writes back to Greenhouse.
- A
Tablessetting that only picks which tables also get a ready-made description (blank means none); it does not limit what the tools can list or query, so limit the connection or the account it signs in with.
Requirements
Java and Maven — clone the repo and run mvn clean install to produce CDataMCP-jar-with-dependencies.jar. Separately install the CData JDBC Driver for Greenhouse and license it (run the driver jar with --license and enter your name, email and either TRIAL or a license key). Then build a connection string with the driver's connection-string utility and write a properties file with Prefix=greenhouse, DriverClass=cdata.jdbc.greenhouse.GreenhouseDriver, the path to the driver jar and that JDBC URL. Point your client at java, the server jar and the properties file. The server speaks stdio, so the client has to run on the same machine. MIT licensed.
Setup effort
Build from source — clone the repository, then point your client at the binary
USE CASES
01
Answering a question about Greenhouse without building a report
Ask in plain language; the assistant lists tables, checks the columns it needs, then runs one SELECT and reads the CSV back.
02
Exploring a schema you have never seen
Start from the table list, drill into columns for the two or three that look relevant, and only then write a query — much cheaper than guessing field names.
3 tools
Listed in project docs. Your client may show a different list.
greenhouse_get_columnsgreenhouse_run_queryBEFORE YOU INSTALL
What to know first
Tools you get
3 tools named in the project's documentation. Tool names take the Prefix in the .prp file, so the example config's names become greenhouse_get_tables, greenhouse_get_columns and greenhouse_run_query. All three are read-only (run_query accepts only SELECT), despite a line promising writes.
Where it runs
Both documented setups run the server on your machine.
INSTALLATION
Connect your client
Clone the repository, then point your client at the binary. This one is built from source rather than installed with a single command.
Server setup
TERMINAL
git clone https://github.com/cdatasoftware/greenhouse-mcp-server-by-cdata.git
cd greenhouse-mcp-server-by-cdata
mvn clean installDownload and install the CData JDBC Driver for Greenhouse from https://www.cdata.com/drivers/greenhouse/download/jdbc
License the driver: java -jar cdata.jdbc.greenhouse.jar --license (enter name, email, and "TRIAL" or a license key)
Run java -jar cdata.jdbc.greenhouse.jar with no arguments to open the Connection String utility, test the connection, copy the connection string
Write a .prp file (e.g. greenhouse.prp) containing Prefix=greenhouse, ServerName, ServerVersion, DriverPath, DriverClass=cdata.jdbc.greenhouse.GreenhouseDriver, JdbcUrl, Tables
Tables only picks which tables also get a ready-made description for your assistant, and empty means none; the server can still list and query every table your connection string reaches, so to keep data out of reach, limit that connection or the account it signs in with.
Add the mcpServers entry below to claude_desktop_config.json, pointing java at CDataMCP-jar-with-dependencies.jar and at the .prp file, then fully quit and reopen the client
MCP Clients
Claude Desktop — macOS/Linux, Windows
JSON
{
"mcpServers": {
"greenhouse": {
"command": "<YOUR_PATH_TO_JAVA>",
"args": [
"-jar",
"<YOUR_PATH_TO_CDATA_MCP_JAR_WITH>",
"<YOUR_PATH_TO_GREENHOUSE>"
]
}
}
}