docusign_get_tablesLABSCO SUMMARY
SQL over DocuSign, without anyone writing SQL.
The three-call shape — list tables, list columns, run a SELECT — is what lets a model find its own way to an answer. It can query every table the connection can see; the Tables setting in the property file does not narrow that, so limit the connection or the account it signs in with. This build is read-only by design; CData points at their commercial server for write access.
What it is
A read-only MCP server that puts DocuSign behind a relational interface. It wraps the CData JDBC Driver for DocuSign, which models the service as SQL tables, and exposes table discovery plus a SELECT runner — so a natural-language question becomes a query the model composes and runs.
What you get
docusign_get_tables— list the tables the driver exposes for the data source, returned as CSV with a header rowdocusign_get_columns— list the columns on one table, so the model knows what it can select before it writes a querydocusign_run_query— execute a SQL SELECT and return the rows- The tool prefix is whatever you set as
Prefixin the connection property file, so several CData servers can sit in one client without colliding
Requirements
A licensed CData JDBC Driver for DocuSign (a trial key works) and DocuSign credentials — OAuth authentication happens in your browser while building the connection string. Java, plus Maven 3 to build CDataMCP-jar-with-dependencies.jar with mvn clean install. You then write a .prp file naming DriverPath, DriverClass, JdbcUrl and an optional Tables list, and the client launches java -jar against the jar and that file. It speaks stdio, so the server must run on the same machine as the client.
Setup effort
Build from source — clone the repository, then point your client at the binary
USE CASES
Answering a question about envelopes without the console
Let the assistant discover the tables, pick the right columns and run a SELECT to get the answer.
Exploring a schema you do not know
List the available tables, then read one table's columns before asking for the rows.
3 tools
Listed in project docs. Your client may show a different list.
docusign_get_columnsdocusign_run_queryBEFORE YOU INSTALL
What to know first
Tools you get
3 tools named in the project's documentation. Tool names take the Prefix value from your .prp file, so with Prefix=docusign they are docusign_get_tables, docusign_get_columns and docusign_run_query. All three only read; write access is CData's separate paid MCP Server.
Where it runs
The documented client runs 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
git clone https://github.com/cdatasoftware/docusign-mcp-server-by-cdata.git
cd docusign-mcp-server-by-cdata
mvn clean installDownload and install the CData JDBC Driver for DocuSign from https://www.cdata.com/drivers/docusign/download/jdbc
License the driver: in the driver's lib folder run java -jar cdata.jdbc.docusign.jar --license, then enter name, email and "TRIAL" (or your license key)
Run java -jar cdata.jdbc.docusign.jar to open the Connection String utility, test the connection and copy the connection string
Create a .prp file with Prefix / ServerName / ServerVersion / DriverPath / DriverClass=cdata.jdbc.docusign.DocusignDriver / 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 and restart the client
MCP Clients
Claude Desktop — macOS / Linux
{
"mcpServers": {
"docusign": {
"command": "<YOUR_PATH_TO_JAVA>",
"args": [
"-jar",
"<YOUR_PATH_TO_CDATA_MCP_JAR_WITH>",
"<YOUR_PATH_TO_DOCUSIGN>"
]
}
}
}