MagicPod offers two ways to create and edit test cases using AI: the “MagicPod MCP Server” and “MagicPod Autopilot.” This page introduces the differences between the two and provides guidelines for when to use each.
Table of Contents
- Comparison Table
- Features of Each
- Guidelines for When to Use Each
- Additional Information
- Related Pages
1. Comparison Table
Both can be used to create and edit test cases using natural language. The differences in functionality are shown in the table below.
| Aspect | MagicPod MCP Server | MagicPod Autopilot |
|---|---|---|
| Where it is used | AI agents (such as Claude and Cursor) | MagicPod’s test case editor |
| AI credits | Not consumed | Consumed |
| File import | You can import files such as Excel files and give instructions based on them | Not supported |
| Scope of operations | Editing test cases and shared steps, running tests, retrieving results, and investigating failures | Editing and summarizing the open test case, and failure analysis |
2. Features of Each
MagicPod MCP Server
By giving instructions in natural language to an AI agent, you can delegate not only the creation and editing of test cases, but also test execution and investigation of the causes of failures.
Benefits
- You can operate MagicPod from AI agents that you are already familiar with, such as Claude, Cursor, and Cline.
- You can incorporate it into AI systems operated within your organization. For example, MagicPod can be used as part of an automated workflow that identifies and modifies affected test cases based on documentation describing specification changes, or has AI investigate CI failures every morning.
- You can create and modify test cases while referencing other tools connected to the AI agent, such as Notion and Slack.
- You can request the creation and modification of multiple test cases, organization of shared steps, test execution, and result verification all at once.
- It does not consume MagicPod AI credits (tokens on the AI agent side are used).
- You can import Excel test procedure documents and specifications to create and modify multiple test cases at once.
- It also supports the creation, editing, and organization of shared steps.
Notes
- You need to prepare an AI agent and set up the MCP server.
MagicPod Autopilot
This is an AI test automation agent used from the chat in the test case editor. You can add and modify steps using natural language while viewing the screen.
Benefits
- You can start using it immediately from the editor without any additional setup (enabling generative AI features is required).
- You can make gradual adjustments while checking the results.
- It can also be used to summarize test cases and analyze failures.
Notes
- It consumes AI credits.
- You cannot edit existing shared steps.
3. Guidelines for When to Use Each
If you are unsure which to use, use the “scale of what you want to do” as a guideline.
MagicPod MCP Server is suitable when you want to create multiple test cases based on specifications or automate the entire process through test execution and failure investigation. On the other hand, MagicPod Autopilot is suitable when you want to make quick fixes while viewing the results in the editor or when you want to try it first without any setup.
The following provides guidelines for each situation.
| When | Recommended |
|---|---|
| When you want to create multiple test cases from Excel specifications or procedure documents | MagicPod MCP Server |
| When you want to automate the process from investigating failed tests through modifying and rerunning them | MagicPod MCP Server |
| When you want to operate without using AI credits | MagicPod MCP Server |
| When you want to organize and edit multiple shared steps at once | MagicPod MCP Server |
| When you want to try AI-powered test creation first without any setup | MagicPod Autopilot |
| When you want to quickly modify part of a test while viewing the screen | MagicPod Autopilot |
You can also use the two together. For example, you can create the overall structure of a test case with the MagicPod MCP Server and then use MagicPod Autopilot in the editor to fine-tune the details.
4. Additional Information
- For neither feature is the content you enter used to train generative AI.
- Generative AI output may not always be as intended. Please review the content before using it.