Skip to main content

​
IBM i MCP Server

A production-ready Model Context Protocol (MCP) server that enables AI agents to securely access IBM i database operations through YAML-configured SQL tools. Purpose-built for enterprise IBM i environments with comprehensive authentication, observability, and security features.

​
Quick Start

Get Started

Set up your server and connect AI agents in 15 minutes

Build Agents

Create specialized IBM i agents using agno and filtered toolsets

​
Key Features

​
🔌 Core Infrastructure

  • MCP Server: Supports stdio and streamable HTTP transport with comprehensive tooling
  • Observability: Built-in OpenTelemetry tracing, metrics, and performance monitoring
  • Type Safety & Security: TypeScript with Zod validation, sanitization, and auth middleware
  • Error Handling: Consistent error categorization, detailed logging, and centralized handling

​
🚀 Production Ready

  • Production Utilities: Logging, error handling, rate limiting, and request context tracking
  • Performance Metrics: Automatic measurement of execution time and payload size
  • Integration Testing: Vitest-based testing with coverage reporting and fast feedback
  • Interaction Logging: Full traceability with dedicated logging for all LLM interactions

​
🤖 Agent Development

  • Agent Ready: Developer cheatsheet and tools tailored for LLM coding agents
  • Services Integration: Reusable modules for LLM and data storage with examples
  • Utility Scripts: Build management, permissions, directory trees, and OpenAPI specs
  • Documentation: Comprehensive guides, JSDoc comments, and API references

​
Documentation

SQL Tools Guide

Create and configure YAML-based SQL tools with parameters and validation

Server Architecture

Understand the server design and component interactions

Configuration

Environment variables and deployment configuration options

​
For IBM i Professionals

Prerequisites: This server requires appropriate IBM i authorities and database access. Ensure your user profile has necessary permissions for intended operations.

​
Common Workflows

Start with the MCP Overview to understand how the Model Context Protocol connects AI agents to IBM i systems.
Follow the Quick Start Guide to get your server running and connected to agents in 15 minutes.
Learn the YAML SQL Tools system to create database operations without writing code.
Explore agent development patterns using agno and FilteredMCPTools for IBM i specialization.

​
Resources

GitHub Repository

Source code, issues, and community contributions

API Reference

Complete API documentation for authentication and endpoints