Your First Project

Get Mibo running against your AI system: create a project, connect your system, and verify the connection.

Note

Head to mibo-ai.com to create your account and follow along.

Create a project

A project is your top-level workspace in Mibo. All your platforms, tests, and results live inside it.

From the dashboard, click New Project. Give it a name and an optional description. We recommend naming it after the AI system you’re testing, for example, “Customer Support Agent” or “Sales Assistant.”

Tip

Write a detailed project description. The Test Architect AI uses your project name and description as context when generating test cases. A good description leads to better, more accurate tests. For example: “A customer support agent that handles billing inquiries, account changes, and product returns for an e-commerce platform.”

One project per AI system. If you have multiple agents or systems, create a separate project for each. This keeps your results clean and your metrics accurate.

Add a platform

A “platform” is the connection between Mibo and your AI system. It tells Mibo where to send test inputs and how to read the responses. Each project can have multiple platforms, which is useful for testing different environments (staging vs. production) or different versions of the same system.

Go to your project, click Add Platform, and choose the type that matches your system:

• Custom API — any system with its own endpoint, whether it’s built with LangChain, your own code, or any other framework.
• Flowise — chatflows and agentflows built with Flowise.
• n8n — AI workflows with webhook triggers built in n8n.

Custom API setup Full guide for Custom API platforms.

Flowise setup Full guide for Flowise platforms.

n8n setup Full guide for n8n platforms.

Tip

Write a clear platform description too. Like the project description, the Test Architect uses the platform description to understand what your system does. Be specific about what this platform handles: “Production Flowise agentflow that handles restaurant reservations, cancellations, and menu inquiries.”

What every platform needs

Regardless of platform type, you’ll always configure:

• A URL: the web address where your AI system receives inputs.
• Authentication (if required): a token, API key, or custom headers that your system expects.
• A message template: the JSON structure Mibo sends to your system, using {input} as a placeholder for the test input.

Note

Platform configuration (URL, authentication, message template) is used for active testing. If you only plan to use passive testing — where your system sends traces to Mibo instead of the other way around — you can set up a platform without connection details and add them later.

Test your connection

After filling in your platform details, click Test Connection to verify everything is working. Mibo will send a quick test input and show you whether it got a valid response.

If the test fails, check these common issues:

• Wrong URL: double-check the address and make sure there are no typos.
• Missing authentication: if your system requires a key or token, make sure it’s entered correctly.
• System not running: verify that your AI system is deployed and accessible from the internet.

Tip

Run a single test case first to verify your connection before running the full suite.

What’s next

Your project is set up and your platform is connected. Next, let’s create test cases.

Creating Test Cases Use the Test Architect to generate test cases from plain language descriptions.