What can I do with the OneLogin MCP server?

You can manage users, applications, and roles in your OneLogin instance via API calls. This includes creating users, assigning apps to users, querying role memberships, and automating identity workflows programmatically.

How do I authenticate this MCP server?

Set three environment variables: ONELOGIN_URL (your instance URL), ONELOGIN_CLIENT_ID, and ONELOGIN_CLIENT_SECRET. These OAuth2 credentials are generated in your OneLogin admin console under API credentials.

Do I need a OneLogin account to use this server?

Yes, you need an active OneLogin instance and admin access to generate OAuth2 API credentials. The server communicates with your OneLogin tenant via the official OneLogin API.

What's the difference between this and calling the OneLogin API directly?

This MCP server wraps the OneLogin API for use in AI agents and LLM-powered tools, making it easier to integrate identity management into agentic workflows without manual HTTP request handling.

Can I use this to sync users from OneLogin to another system?

Yes, you can query user data from OneLogin via this server and then use that data in downstream tools. However, the server itself is read/write for OneLogin only—you'd need additional tools to push data elsewhere.

MCP ServerSTDIOOfficialv1.0.18

Onelogin MCP Server

Manage OneLogin identity and access control programmatically—create users, assign apps and roles, and query authentication data. Built for developers automating IAM workflows in AI agents and scripts.

io.github.onelogin/onelogin-mcp

Try MCP with AI models View source

Hosted URL

Local install

Transport

STDIO

Auth

3 required env vars

What the Onelogin MCP server does

How models use it and what it is built for.

Connect to Onelogin

Local install — runs as a subprocess.

npx @onelogin/onelogin-mcp@1.0.18

Environment variables

Configuration this server reads at startup.

ONELOGIN_URLRequired
Your OneLogin instance URL (e.g. https://mycompany.onelogin.com)
ONELOGIN_CLIENT_IDRequiredSecret
OneLogin OAuth2 API client ID
ONELOGIN_CLIENT_SECRETRequiredSecret
OneLogin OAuth2 API client secret

Resources

Where to find authoritative docs and source for Onelogin.

Source repository

Example prompts for Onelogin

Paste any of these into Agent Studio after connecting Onelogin.

List all users in my OneLogin instance with their assigned roles
Create a new user account with email and assign them to the Engineering app
Query which apps are assigned to a specific user role
Explain the authentication flow and required OAuth2 credentials for this MCP server

Onelogin MCP server — FAQ

Common questions about connecting and running Onelogin.

What can I do with the OneLogin MCP server?
You can manage users, applications, and roles in your OneLogin instance via API calls. This includes creating users, assigning apps to users, querying role memberships, and automating identity workflows programmatically.
How do I authenticate this MCP server?
Set three environment variables: ONELOGIN_URL (your instance URL), ONELOGIN_CLIENT_ID, and ONELOGIN_CLIENT_SECRET. These OAuth2 credentials are generated in your OneLogin admin console under API credentials.
Do I need a OneLogin account to use this server?
Yes, you need an active OneLogin instance and admin access to generate OAuth2 API credentials. The server communicates with your OneLogin tenant via the official OneLogin API.
What's the difference between this and calling the OneLogin API directly?
This MCP server wraps the OneLogin API for use in AI agents and LLM-powered tools, making it easier to integrate identity management into agentic workflows without manual HTTP request handling.
Can I use this to sync users from OneLogin to another system?
Yes, you can query user data from OneLogin via this server and then use that data in downstream tools. However, the server itself is read/write for OneLogin only—you'd need additional tools to push data elsewhere.

Skip the local setup — run MCP in your browser

MCP Playground runs 10,000+ hosted MCP servers — GitHub, Linear, Notion, Stripe, Sentry and more — across Claude, GPT, Gemini, DeepSeek and 60+ AI models. Compare model answers side-by-side, save agent presets, share runs. Zero install.

Open Agent Studio