
IBM Maximo
io.github.markusvankempenPublished by GitHub account markusvankempenv1.1.0Updated Oct 24, 2026
Connect AI assistants to IBM Maximo for API discovery, OSLC querying, and Carbon UI generation.
Overview
Connects an AI assistant to an IBM Maximo instance for API discovery, schema inspection, OSLC queries, record creation and updates, and Carbon UI generation.
- What you can do
- Discover and inspect Maximo APIs: list object structures and get field definitions for each one, plus check instance connectivity and server details. Query live data: run OSLC REST queries against the Maximo instance. Generate UI: render Carbon Design tables and record detail views. Write data: create records in any object structure, partially update records by ID, and run business actions such as status changes and approvals.
- When to use it
- Useful when working with an IBM Maximo environment and you want an assistant to look up object structures and fields, test OSLC queries against live data, scaffold Carbon tables or dashboards, or create and update work orders, assets and service requests through prompts.
- Requirements
- Runs locally over stdio via the npm package maximo-mcp-server, so Node.js v18 or higher is needed. Requires a reachable Maximo instance and the environment variables MAXIMO_URL (REST API base URL), MAXIMO_HOST (host URL) and MAXIMO_API_KEY. An OpenAPI schema file can optionally be downloaded for offline schema lookups.
Installation
In SourceWeft
- Open IBM Maximo in the dashboard and add it to a workspace.
- Enable the server for the chats that should use its tools.
Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.
Other MCP clients
Follow the launch instructions in the repository.
README
🏭 Maximo MCP Server
AI-Powered Development for IBM Maximo
[Node.js] [MCP] [Maximo] [License] [Version]
Transform your Maximo development workflow with AI-driven schema discovery, live data querying, intelligent code generation, and full write-back capabilities.
Author: Markus van Kempen
Email: [email protected] | [email protected]
Date: 16 September 2026
🎯 What is This?
The Maximo MCP Server is a Model Context Protocol server that connects AI assistants (like Antigravity, Cursor, or VS Code Copilot) directly to your IBM Maximo environment. Instead of manually copying API documentation, the AI can:
📚 Documentation
Core Guides
French Translations
Word Documents
All guides are also available in .docx format in the docs/ folder for offline reading and sharing.
⚡ Quick Start
Prerequisites
- Node.js v18 or higher
- Maximo API Key with read access
- AI IDE with MCP support (Antigravity, Cursor, VS Code + Continue)
Installation
Installation
Method 1: Run directly with npx (Recommended)
Method 2: Clone from Source
Environment Configuration
Edit the .env file with your Maximo credentials:
Download the OpenAPI Schema (Recommended)
The OpenAPI schema file enables offline schema lookups for faster AI responses:
Alternatively, download via Swagger UI at: https://your-host/maximo/oslc/oas/api.html (Click "Explore" or "Download")
Method 3: Direct Browser Download (Manual)
If curl fails (e.g., due to SSL/network errors), you can manually download the file:
-
Open this URL in your browser:
https://[YOUR_MAXIMO_HOST]/maximo/oslc/oas/api(Replace[YOUR_MAXIMO_HOST]with your actual server address) -
You may be prompted to log in to Maximo.
-
Once the JSON loads, right-click the page and select "Save Page As...".
-
Save the file as
maximo_openapi.jsonin your project root folder.
Note: This file is ~12MB and contains all Object Structure definitions for your Maximo instance.
IDE Configuration
VS Code with GitHub Copilot (Recommended)
Option 1: Install from the MCP Server Gallery
- Enable
chat.mcp.gallery.enabledin VS Code settings - Open the Extensions view (
⇧⌘X) - Type
@mcp maximoin the search field - Click Install to add the Maximo MCP server
Option 2: Add manually via mcp.json
- Open the Command Palette (
⇧⌘P) → MCP: Open Workspace Folder Configuration - Add the following configuration:
- VS Code will prompt you for your Maximo credentials when the server starts.
💡 Tip: This project includes a
.vscode/mcp.jsonfile. If you clone the repo, VS Code will auto-detect the MCP server configuration.
Google Antigravity (Manual Setup Required)
⚠️ Note: The Antigravity MCP Store is curated and does not auto-discover servers from the registry. You must add this server manually.
- Open Antigravity
- Click "..." dropdown at the top of the Agent panel
- Select "MCP Servers" → "Manage MCP Servers" → "View raw config"
- Add to your
mcp_config.json:
- Save and click Refresh
Cursor / Claude Desktop
Edit with your Maximo credentials:
Verify Connection
In your AI IDE, ask:
"Is the Maximo MCP server connected?"
The AI will call get_instance_details and confirm connectivity.
🎬 Live Demo
Asset Manager Application
We built a complete Maximo Asset Manager web application using only natural language prompts and the MCP server.
50 assets loaded with real-time filtering and search
Demo Features
🎥 Screen Recording
A complete video demonstration is available: assets_demo_recording.webp
Try It Yourself
🛠 Available MCP Tools
The server exposes 9 tools to the AI — 6 read tools and 3 write/CRUD tools:
Read Tools
Write Tools (CRUD)
⚠️ Write tools modify live data. Use a read-only API key for exploration; only enable write access for known workflows.
💡 Use Cases
1. Generate API Calls
"Get me the last 10 approved work orders from BEDFORD site"
The AI calls get_schema_details(MXWO), understands the fields, and generates:
2. Generate Python Scripts
"Write a Python script to export all Priority 1 work orders to CSV"
3. Generate SQL Queries
"Write SQL to find overdue work orders"
4. Build Complete Applications
"Create an HTML dashboard to display assets"
Result: A complete web application with:
- Dark theme with glassmorphism
- Search and filter functionality
- Interactive detail panels
- Pre-loaded data from Maximo
See the Asset Manager Case Study for the full walkthrough.
5. Create & Update Records (CRUD)
"Create a corrective maintenance work order for the BEDFORD site, priority 1, description 'Pump failure inspection'."
The AI calls get_schema_details(MXWO) to confirm field names, then create_record:
"Now approve work order 1025 with memo 'Reviewed and approved'."
See the full guide: Maximo MCP Server Guide — CRUD Workflows
🧩 VS Code Extensions
This project includes two VS Code extensions for interactive Maximo API development — no AI agent required.
Maximo API Explorer
A full-featured VS Code extension for discovering, testing, and generating code for Maximo REST APIs.
[Maximo API Explorer — Sidebar]
Connected to a live Maximo instance showing Object Structures (MXWO, MXSR, MXASSET, etc.) and API Endpoints.
[Maximo API Explorer — API Tester]
OSLC Query Builder with live JSON response (200 OK, 20 records from MXWO).
Quick Start
See the full guide: Maximo API Explorer Guide
Maximo Cursor Explorer
A fork of the API Explorer optimized for Cursor's AI features:
📁 Project Structure
🔒 Security Best Practices
🤝 Contributing
Contributions are welcome! Please read our contributing guidelines before submitting PRs.
- Fork the repository
- Create a feature branch (
git checkout -b feature/amazing-feature) - Commit your changes (
git commit -m 'Add amazing feature') - Push to the branch (
git push origin feature/amazing-feature) - Open a Pull Request
📄 License
This project is licensed under the Apache License 2.0 - see the LICENSE file for details.
🙏 Acknowledgments
- Model Context Protocol for the MCP specification
- IBM Maximo for the enterprise asset management platform
- Carbon Design System for the UI components
Source: README.md at commit 0d02315
Tools
0Version history
1- v1.1.0LatestSep 16, 2026
Score
50% of the applicable checks met
How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.
- Reached in the last 30 days25 pts
Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.
- Auth declared10 pts
Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.
- Tools listed15 pts
Answer tools/list without a credential, or ship a package that starts in a sandbox without one.
- DescriptionRequired15 pts
- Logo10 pts
- Documentation15 pts
- License10 pts