- MOCK
- DATA
- VALIDATION
- HUMAN
Short answer
How can I make my AI act like a mock API?
You give your AI a webhook URL and tell it what status code and body to return. The AI then configures the endpoint to respond exactly how you want. You'll get a controlled environment to test your webhooks and integrations.
The testing outcomes
Where your tests land
Once the connection is live, your AI manages these four states.
MOCK
Custom status codes
The AI sets specific HTTP codes like 200 or 500. This lets you test how your app handles success and failure.
DATA
Specific response bodies
You define the exact JSON or text payload. Your AI ensures the endpoint returns that specific data when hit.
VALIDATION
Endpoint verification
The AI uses the configured URL to act as a target. You can verify that your outgoing webhooks hit the right place.
HUMAN
Manual adjustments
You decide when to change the response logic. The AI only changes what you explicitly tell it to via chat.
The workflow
What your AI does when the request hits the URL.
The AI acts as the logic layer between the incoming request and the response you want to see.
Establish the endpoint
The AI generates a unique URL to act as your target for testing.
create_tokenDefine the response
You tell the AI what status code and body the endpoint should use.
set_responseConfigure the logic
The AI applies your custom status and payload settings to that specific URL.
set_responseExecute the test
The AI waits for the request to arrive and serves the pre-configured response.
set_response
Try it
Copy these to start.
Use these prompts to begin mocking your endpoints.
Starting points
Swap out the URL and the JSON details for your specific test case.
Webhook.site Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_4UGcdultUkkqhC6PeYxs6JG990XLacW5YNGNQO5y/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — Webhook.site capabilities are ready to use.
{
"mcpServers": {
"webhooksite-mcp": {
"url": "https://edge.vinkius.com/vk_preview_4UGcdultUkkqhC6PeYxs6JG990XLacW5YNGNQO5y/mcp"
}
}
}Create a new webhook URL for me to test.
Set the response for https://webhook.site/test-id to a 200 OK with a JSON body of {"status": "success"}.
Configure the endpoint at https://webhook.site/error-test to return a 500 Internal Server Error.
Make the webhook at https://webhook.site/data-test return a 201 Created status with this payload: {"id": 123}.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect Webhook.site once, then ask.
Just click the link to connect. Your credentials stay encrypted, and you can start prompting immediately.
Connect Webhook.site to your AIFAQ
How the mocking works
- 01
Can the AI change the URL after it is created?
No. The AI creates a specific URL using the token tool, and you must use that specific URL for your tests.
- 02
Can the AI see the data I send to the webhook?
No. The AI only sets the response that the webhook sends back to you. It does not intercept or read the incoming request data.
- 03
How do I test a 404 error?
Tell the AI to set the response for your URL to a 404 status code using the response tool.
- 04
Can the AI host a permanent API?
No. This is for simulating responses during testing, not for hosting a production API.
- 05
Does the AI manage my Webhook.site account?
No. The AI only uses the tools provided to create tokens and set responses for your testing sessions.
More questions about Webhook.site? See everything the Webhook.site Connector can do
Connect Webhook.site to Claude, Cursor, ChatGPT & more
