IBM Informix

You can query live IBM Informix data through a read-only MCP server that wraps the CData JDBC driver. This MCP server exposes IBM Informix data as MCP tools, so you can ask natural language questions and receive up-to-date results without writing SQL.

How to use

After you configure and run the MCP server, connect your MCP-enabled AI client to access live IBM Informix data. You can ask questions like the current account list, recent transactions, or open tickets, and the client will retrieve results from Informix through the MCP tools. The server provides tools to list available tables, list columns for a table, and run a SELECT query, returning results in a structured format suitable for natural language querying.

How to install

Prerequisites you need before installation: a Java Runtime Environment (JRE) or JDK, and Maven for building the server.

Step 1: Clone the MCP server repository and enter the project directory.

git clone https://github.com/cdatasoftware/ibm-informix-mcp-server-by-cdata.git
cd ibm-informix-mcp-server-by-cdata

Step 2: Build the MCP server. This creates the runnable JAR with dependencies.

mvn clean install

Step 3: Obtain and license the CData JDBC Driver for IBM Informix. Download from the CData site and install it following the provided installer and guidance for your OS.

Step 4: Configure the JDBC connection and generate a property file for MCP access. The process typically includes running the JAR in connection-string mode to test and then saving the connection as a .prp file.

java -jar cdata.jdbc.informix.jar
# Use the Connection String utility to configure and test
# Copy the resulting JDBC URL to later include in your .prp file

Step 5: Create the MCP connection profile. Create a file such as ibm-informix.prp with the following properties and format, tailored to your environment.

env
Prefix=informix
ServerName=CDataInformix
ServerVersion=1.0
DriverPath=/path/to/cdata.jdbc.informix.jar
DriverClass=cdata.jdbc.informix.InformixDriver
JdbcUrl=jdbc:informix:InitiateOAuth=GETANDREFRESH;
Tables=

Additional configuration and running notes

Claude Desktop users can add the MCP server entry to their config so the client can start the server locally. The server runs on the same machine as the client when using stdio mode.

To start the MCP server from the built JAR using your .prp file, run the following command with the appropriate paths.

java -jar /PATH/TO/CDataMCP-jar-with-dependencies.jar /PATH/TO/ibm-informix.prp

Troubleshooting and tips

If the MCP server does not appear in your client, ensure you have fully quit and restarted the client after configuration. Double-check the path to the JAR and the .prp file, and verify that the JDBC driver is licensed and loaded correctly.

If data cannot be retrieved, confirm the JDBC URL and tables configuration in the .prp file. Re-run the connection string utility to validate the connection.

Notes on tools and usage

The MCP server exposes three tools to interact with IBM Informix data. Each tool is available for use by your MCP client to discover data structure and retrieve results.

Security and licensing considerations

This MCP server relies on the licensed CData JDBC Driver for IBM Informix. Ensure you keep the JDBC driver license valid and comply with the licensing terms of all software involved.

Available tools

informix_get_tables

Retrieves a list of tables available in the IBM Informix data source. Output is CSV with headers on the first line.

informix_get_columns

Retrieves a list of columns for a specified table. Output is CSV with headers on the first line.

informix_run_query

Executes a SQL SELECT query against the IBM Informix data source and returns the results.