# Webhook.site MCP for AI Agents AI Agent Connect > Webhook.site MCP lets you create, manage, and inspect HTTP webhooks directly through your AI client. Use it to debug third-party callbacks, test API integrations, and simulate complex request and response behaviors without having to manually click through a dashboard. ## Overview - **Category:** developer-tools - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_4UGcdultUkkqhC6PeYxs6JG990XLacW5YNGNQO5y/ai-agent-connect - **Tags:** webhook-testing, api-debugging, http-inspector, request-bin, payload-analysis ## Description Webhook.site connects your AI agent to a professional webhook testing environment where you can manage HTTP traffic using natural language. If you're trying to figure out why a payment gateway isn't hitting your endpoint, you usually have to hop between your code, a terminal, and a browser tab just to see the raw data. This Connector takes that friction out of the equation. You can tell your agent to spin up a new webhook URL with a specific alias like stripe-test and then ask it to watch for incoming pings. Once a request hits, you can have your agent list the requests, show you the headers, or even check the JSON payload for a specific field. It makes the manual labor of inspecting traffic disappear. If you need to simulate a specific failure, you can have the agent set a dynamic response with a 500 error code and a custom body to see how your app handles it. It's one of the reasons I appreciate how the Vinkius catalog makes these types of developer tools so accessible. You get to move from wondering why something is broken to seeing the actual error in the payload in a few seconds. You can even manage environment-wide variables to keep state across different tests, ensuring your agent has the context it needs to stay accurate. ## Tools ### create_token Create a new Webhook.site token with a custom URL. It allows you to generate unique endpoints for different services. ### get_token Get details for a specific token. Use this to check the current configuration and status of a webhook. ### list_actions List custom actions for a token. This shows all the automated responses currently configured for your URL. ### list_global_variables List global variables. This lets you see all the environment-wide data your agent is currently tracking. ### list_tokens List Webhook.site tokens. Use this to see all your active URLs in one view. ### update_global_variable Update a global variable. This lets you change shared state values on the fly during a test. ### update_token Update an existing token. Use this to change the alias or default behavior of a webhook URL. ### create_global_variable Create a global variable to store state across different webhook tests. This helps maintain context during complex debugging. ### delete_action Delete a custom action from a token. Use this to clean up your automated responses when they are no longer needed. ### delete_global_variable Delete a global variable from your environment. This keeps your workspace clean of old test data. ### delete_requests Delete multiple requests for a specific token. It's the fastest way to clear your history during heavy testing sessions. ### delete_token Delete a token from your account. Use this to remove old webhook URLs you no longer use. ### execute_action Execute actions for a specific request. This allows you to trigger automated logic on incoming hits. ### get_requests Get requests captured by a token. This lets you see every header and payload that hits your endpoint. ### set_response Set dynamic response for a specific request. This is how you simulate custom API behaviors and error codes. ### update_action Update a custom action. Use this to modify existing automated responses without creating new tokens. ### create_action Create a custom action for a token. This lets you define new automated responses for incoming requests. ## Prompt Examples **Prompt:** ``` I need to test my new webhook. Can you make a URL for 'auth-callback' and tell me when it gets hit? ``` **Response:** ``` I've created a new webhook for you. **URL:** https://webhook.site/auth-callback **Status:** Active I'll monitor this for you. Just send a request to that URL, and I'll let you know the details of the payload, headers, and the source IP as soon as it arrives. ``` **Prompt:** ``` Show me the last 3 requests that came into my 'payment-test' webhook. ``` **Response:** ``` I found 3 recent requests for the **payment-test** token: | Time | Method | Status | Payload Summary | | :--- | :--- | :--- | :--- | | 12:01 PM | POST | 200 | `{"amount": 50, "currency": "USD"}` | | 12:02 PM | POST | 200 | `{"amount": 25, "currency": "USD"}` | | 12:05 PM | POST | 400 | `{"error": "invalid_card"}` | Would you like me to show you the full headers for the error request? ``` **Prompt:** ``` Make the 'stripe-test' webhook return a 503 Service Unavailable error with a 'Maintenance' message. ``` **Response:** ``` I've updated the **stripe-test** token configuration. **New Response Settings:** * **Status Code:** 503 * **Body:** `{"message": "Maintenance"}` Any incoming requests to this URL will now receive this response. This is great for testing your app's retry logic. ``` ## Capabilities ### Spin up custom webhook URLs Generate unique URLs with specific aliases and default behaviors using create_token. ### View incoming request data Retrieve detailed headers, query parameters, and raw payloads using get_requests. ### Set custom API responses Simulate specific API behaviors by defining status codes and bodies with set_response. ### Manage environment variables Store and update state across different webhook tests using create_global_variable. ### Automate request actions Trigger specific logic on captured requests automatically with execute_action. ## Use Cases ### Debugging a Stripe payment callback A developer asks the agent to create a token for 'stripe-test' and wait for a webhook. Once the hit arrives, the agent lists the requests and shows the JSON payload to find the missing signature. ### Simulating a 500 error from a third party A QA engineer needs to see how the app handles a crash. They tell the agent to use set_response to return a 500 status code and a custom error message for all incoming pings. ### Managing environment-wide variables A dev uses create_global_variable to store a shared secret across multiple webhook tests, ensuring the agent can use the same value in different scenarios without re-typing it. ### Cleaning up test data After a long testing session, a dev asks the agent to delete_requests for a specific token to wipe the history and start fresh for the next round of debugging. ## Benefits - Skip manual dashboard navigation by creating tokens and viewing requests through your AI client using create_token. - Simulate complex API behaviors by using set_response to define specific status codes and bodies for your tests. - Maintain state across multiple tests using create_global_variable to store environment data for your agent. - Clear clutter quickly with delete_requests to keep your webhook history clean during heavy testing sessions. - Automate your testing pipeline by using execute_action to trigger specific logic on incoming hits automatically. ## How It Works The bottom line is you get a way to debug and simulate webhooks using only natural language. 1. Add your Webhook.site API Key to your environment variables. 2. Connect the Connector to your preferred AI client like Claude or Cursor. 3. Ask your agent to create a token and monitor incoming traffic. ## Frequently Asked Questions **Can I use Webhook.site MCP to see what data my app is sending?** Yes. You can use it to capture and inspect any outgoing HTTP request. Your agent can pull the headers and the raw JSON payload so you can see exactly what your app is doing. **How do I use Webhook.site MCP to test a failed API call?** You can use it to simulate failures easily. Just tell your agent to set a dynamic response with a specific error code, like a 404 or 500, and a custom message. **Is Webhook.site MCP good for debugging Stripe or Twilio webhooks?** It's perfect for that. You can create a unique URL for each integration, watch the incoming data in real-time, and verify that the payloads match what you're expecting from those providers. **Can Webhook.site MCP help me manage multiple test URLs at once?** Yes, it can. Your agent can list all your active tokens, create new ones with specific aliases, and even delete old ones to keep your workspace organized. **Does Webhook.site MCP support global variables for my tests?** Yes, it does. You can create and update global variables to store state or configuration data that your agent can access across different webhook test runs. **Can I use Webhook.site MCP to clear out my request history?** You can. If your logs get too cluttered during a long debugging session, just ask your agent to delete the requests for a specific token to start with a clean slate. **How can I create a temporary webhook URL that expires after one hour?** Use the `create_token` tool and set the `expiry` parameter to 3600 seconds. You can also add an `alias` to make the URL easier to identify. **Can I see the headers and body of the requests sent to my webhook?** Yes! Use the `get_requests` tool with your Token ID. It will return a list of captured requests including full headers, query strings, and the raw payload content. **Is it possible to make the webhook return a specific JSON response?** Absolutely. Use the `set_response` tool to define a custom `content` (base64 encoded), `status` code, and JSON `headers` for any specific request received by your token.