Connect Slack, Zapier and Your Deploy Pipeline

Alert emails are fine until your team lives in Slack, your tasks live in Trello, and your deploys run through a script. Then an email is one more inbox to watch, and a broken page waits until somebody opens it.

WebChange Detector can push its results into the tools you already use, and it can be started by them. This guide covers the three ways in, from no code to a little code:

  • Zapier: connect WebChange Detector to Slack, Trello, Google Sheets and thousands of other apps without writing code.
  • Webhooks: WebChange Detector sends each event as JSON to a URL of your choice, signed so your server knows it really came from us.
  • REST API: start checks from your deploy script, for example “Run Monitoring now” right after a push to live.

Webhooks and Zapier live on the Automations tab of the Web App.

Zapier: Slack, Trello and more without code

Pick Zapier when you want the result in another app and nobody on the team wants to maintain a server. WebChange Detector has its own app on Zapier, and the Automations tab has the Zapier workflow builder built in.

  1. Open the Automations tab and scroll to the card Zapier Integration.
  2. Search for the app you want to connect, for example Slack, and follow the steps. The builder is pre-filled with your email address, so signing up or logging in to Zapier takes a moment.
  3. Connect your WebChange Detector account when Zapier asks for it, choose the action in the other app (post a message, create a card, add a row) and switch the Zap on.

A few ideas that work well for agencies:

  • A message in the client’s Slack channel when a check finds changes on their site.
  • A Trello or Asana card per change, so somebody owns the fix.
  • A Google Sheets row per finished check run, as a running log for your monthly client report.

Add a filter step in Zapier (for example only one client’s website) so the channel does not get noisy.

Webhooks: send events to your own endpoint

Pick a webhook when you have your own system: a ticket tool with an API, a CI pipeline, a custom client dashboard. WebChange Detector sends a POST request with JSON to your URL whenever the event happens.

The four events

EventFires whenTypical use
New Change DetectedA comparison finds a new change on a pageInstant alert or a ticket per change
Batch CompletedAll screenshots and comparisons of one check run are doneOne summary per run instead of one message per page
Screenshot CompletedA single screenshot was takenProgress tracking in your own tools
Screenshot FailedA screenshot could not be taken, for example because of a timeout or a blockAlert when a site is down or blocks the screenshots

“New Change Detected” also tells you what started the check: a scheduled Monitoring run, a saved page, “Run Monitoring now”, an On-Demand Check or an Auto Update Check. That lets your endpoint treat a change after a deploy differently from a change found at night.

Create and test a webhook

  1. On the Automations tab, click Create Webhook in the Webhooks card.
  2. Choose the Event Type and enter the Webhook URL, the https:// address of your endpoint.
  3. Click Create. The webhook shows up in the list below.
  4. Open it and click Test. WebChange Detector sends a sample delivery to your URL, so you see right away whether your endpoint answers.

Each webhook has Edit and Delete next to Test. One webhook covers one event; to get two events, or one event at two URLs, create two webhooks. Below the list, Webhook Event Documentation explains every event with an example payload, so your developer knows exactly which fields arrive.

Creating a webhook: choose the event, enter the URL, and the new webhook appears with its Test button.

Tip: you can not point a webhook straight at a Slack incoming webhook, because Slack expects its own message format. For Slack, use Zapier, or a small script of your own in between.

Signed deliveries: check that it really came from us

A webhook URL is public. Anybody who knows it could send fake “all good” messages to it. That is why every delivery carries a Signature header: an HMAC-SHA256 hash of the request body, made with your account’s own webhook secret. Your endpoint calculates the same hash with the same secret and only accepts the delivery when both match.

You get the secret from the REST API: the account endpoint (GET /api/v2/account) returns it as webhook_secret. Keep it on your server, never in front-end code. If your endpoint is down, failed deliveries are retried. One exception: the sample sent by the Test button carries no signature, so expect that while you set things up.

REST API: start checks from your deploy pipeline

Everything the dashboard does runs through the WebChange Detector REST API: websites, groups, URLs, screenshots, Checks, webhooks and subaccounts. Agencies use it to start checks from a deploy script or to show results in their own client portal. Changes to the API only add things, so a working integration does not break after an update.

  • Base URL: https://api.webchangedetector.com/api/v2/
  • Login: send your API token as Authorization: Bearer YOUR_API_TOKEN. You find the token on the Dashboard tab under Show API Token. A subaccount’s token is on its row in the Subaccounts tab.
  • Reference: every endpoint with examples is in the API documentation.

Run Monitoring now as the last step of a deploy

Some hosts, WP Engine for example, push staging to live as a full copy. Nothing gets “saved” on the live site, so the check on page save never fires, and the next scheduled Monitoring run may be hours away. One API call at the end of your deploy script checks every monitored page of the website right away. Why this matters and how to do the same with one click is in Run Monitoring now after a deploy.

Step 1: find the Monitoring group of the website (once). Each website has a Monitoring group. List them and note the id of the one with your website’s name:

curl -s "https://api.webchangedetector.com/api/v2/groups?monitoring=1" \
  -H "Authorization: Bearer YOUR_API_TOKEN" \
  -H "Accept: application/json"

Step 2: start the run from your deploy script. scope all checks every active page of the group:

curl -s -X POST "https://api.webchangedetector.com/api/v2/monitoring/trigger" \
  -H "Authorization: Bearer YOUR_API_TOKEN" \
  -H "Accept: application/json" \
  -H "Content-Type: application/json" \
  -d '{"group_id": "YOUR_GROUP_ID", "type": "manual_trigger", "scope": "all"}'

To check only the pages you changed, leave out scope and send them as a list instead: "urls": [{"url": "https://example.com/pricing/"}].

What the answer means:

StatusMeaning
202The run started. The answer contains the batch_id of the run.
409, reason already_runningThese pages are being checked right now. Nothing is charged twice.
402, reason not_enough_creditsYour plan does not have enough checks left for this run. Nothing was started.
422, reason trigger_not_enabledMonitoring is switched off for this website. Switch it on in the dashboard first.

Two practical tips:

  • Purge the cache first. The run starts the moment the call arrives. If your deploy clears a page cache or a CDN, put the call after that step, otherwise the screenshots may show the old version.
  • Get the result where you need it. The run behaves like any Monitoring run: AI verification, alert emails and webhooks included. In the Checks tab it is labelled “Manual Trigger”. Combine it with a “Batch Completed” webhook and your pipeline knows when the check is done.

WebChange Detector Newsletter

News, feature Updates, and insights.
No spam. Unsubscribe anytime.