LLM Webhooks Testing locally Helicone

This is a short guide on how to test webhooks locally.

How to test webhooks locally

The idea is that we are going to use a tool like ngrok to create a secure tunnel to your local machine so that Helicone can send events to your local server.

Create your local worker server

Here is a simple python example using FastAPI:

python3 -m pip install fastapi uvicorn

Create a file called main.py and add the following code:

from fastapi import FastAPI, Request

app = FastAPI()

@app.post("/")
async def webhook(request: Request):
    return {"message": "Webhook received"}

Run the server using:

python3 -m uvicorn server:app --reload --port 9393

you should see something like this…

INFO:     Uvicorn running on http://127.0.0.1:9393 (Press CTRL+C to quit)
INFO:     Started reloader process [20505] using StatReload
INFO:     Started server process [20507]
INFO:     Waiting for application startup.
INFO:     Application startup complete.

Setting up ngrok

1. Install ngrok using the official guide.
2. Run ngrok http <port>, where <port> is the port that your local server is running on.

ngrok http 9393

You should see something like this…

Session Status                online
Account                       [e-mail] (Plan: Free)
Version                       3.15.1
Region                        United States (us)
Latency                       63ms
Web Interface                 http://127.0.0.1:4040
Forwarding                    https://9742-103-249-231-120.ngrok-free.app -> http://localhost:9393

Copy the ngrok URL

Add the URL outputed from ngrok to your webhook configuration in the Helicone webhooks page.

in our example it would be: https://9742-103-249-231-120.ngrok-free.app

Webhooks page

For this testing example I added the property filter environment to equal to dev so that it only sends events to this webhook from our development environment.

Send a request through Helicone to receive the webhook

curl --request POST \
  --url https://oai.helicone.ai/v1/chat/completions \
  --header 'Authorization: Bearer $OPENAI_API_KEY' \
  --header 'Helicone-Auth: Bearer $HELICONE_API_KEY' \
  --header 'Helicone-Property-environment: dev' \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "gpt-3.5-turbo",
    "messages": [
        {
            "role": "system",
            "content": "Say Hello!"
        }
    ],
    "temperature": 1,
    "max_tokens": 10
}'

After sending the request you should see the webhook hit your local server.