Skip to main content
Experimental Hoppscotch now offers a set of AI-powered features designed to simplify core stages of the API development lifecycle. These include renaming of API requests, generation of structured request payloads, and scripting capabilities for pre-request logic and test cases.
These AI features are currently in the alpha stage and are exclusively available on Hoppscotch Cloud and our Desktop app. They are NOT supported in the Self-Hosted Edition at this time.

​
Enable AI-powered features

To activate AI-powered features, follow these steps:
  1. Open the Settings page within the Hoppscotch application.
  2. Scroll to the Experiments section.
  3. Toggle the AI Experiments option to enable the features.

​
Rename API Requests

When saving an API request to a collection, click the Modify with AI icon ( .⟡⁺ ) next to the request name field to have the system automatically assign a name based on the request’s context and properties. Alternatively, you can use this feature to rename the pre-saved requests. You can even customize the naming convention (e.g., CamelCase, PascalCase) through the Request Naming Style dropdown located in the Experiments section of the Settings page.

​
Generate Request Body

Dynamically generate structured request payloads:
  1. Click the Modify with AI icon ( .⟡⁺ ) in the Request Body section.
  2. Define constraints or provide input for the desired structure.
  3. Validate the generated body and select Accept Change to populate the request body field.

​
Generate Pre-request Scripts

Generate pre-request scripts to handle necessary setup operations before sending the request:
  1. Click the Modify with AI icon ( .⟡⁺ ) within the Pre-request Script section.
  2. Enter specific logic or conditions required for the pre-request phase.
  3. After reviewing the generated script, click Accept Change to apply it to the pre-request script section of your API request.

​
Generate Test Scripts

Develop test cases for your API workflow:
  1. Click the Modify with AI icon ( .⟡⁺ ) in the Tests section.
  2. Specify the test parameters or expected outcomes.
  3. Select Accept Change to integrate it into the test script editor for execution.

​
Connect an AI agent with the MCP server

The features above run inside the Hoppscotch app. To let an external AI agent work with your collections, environments, and teams, use the Hoppscotch MCP server. It exposes those resources to hosts such as Claude Desktop, Claude Code, and Codex, and can execute HTTP requests on your behalf.
The MCP server is a separate, generally available product. The experimental status and the alpha notice above apply to the in-app AI features, not to it. Unlike those features, it works with both Hoppscotch Cloud and self-hosted instances.

Hoppscotch MCP Server

Install the server, connect your MCP host, and review what each tool can do.