Db2 for i

IBM i MCP Server

What the IBM i MCP Server is: an open-source, IBM-published Model Context Protocol server that lets AI agents and LLMs query Db2 for i data and take action on IBM i using natural language.

The IBM i MCP Server is an open-source project published by IBM that implements the Model Context Protocol (MCP), an open standard that lets AI agents and LLM-powered tools such as Claude, GPT-4, and local LLMs query Db2 for i data, run SQL, inspect system objects, and take defined actions on IBM i, all through natural language rather than hand-written SQL or RPG. It is a connectivity and tooling layer for AI agents, not a machine learning or predictive-modeling feature.

How it works

ProtocolModel Context Protocol (MCP), an open standard for connecting AI agents to external tools and data
Connectivity layerMapepire, a secure, modern Db2 for i connectivity layer
Tool definitionsYAML-defined SQL tools, shared with a companion ibmi CLI (@ibm/ibmi-cli)
StatusBeta; opened October 27, 2025

What it actually does

Natural-language querying

Lets an AI agent answer questions about Db2 for i data, such as current inventory levels or open orders, by translating a plain-language question into the right query behind the scenes.

System object inspection

Gives an MCP-compatible AI client the ability to inspect IBM i system objects and metadata as part of answering a question or completing a task.

Action-taking

Supports defined, tool-scoped actions against IBM i data and objects, not just read-only lookups, depending on which YAML-defined tools are exposed.

What it is not

The IBM i MCP Server is not a machine learning or predictive-modeling capability. It does not train or score models, and it does not perform anomaly detection, demand forecasting, or predictive maintenance on its own. It is an agent-connectivity layer: it lets an external AI agent or LLM query and act on Db2 for i data using natural language, over MCP, with Mapepire handling the underlying database connection.

The IBM i MCP Server is in beta as of its October 2025 opening. Confirm current capability, supported tool definitions, and production-readiness directly against IBM's GitHub repository and documentation before scoping a project around it.