Our webhook receiver went quiet on a Tuesday. No vendor incident. Status page still green. Support said "check the docs." The docs still looked fine until I scrolled to the endpoint table. /v1/hooks was gone. The new path was /v2/webhooks, with a renamed signature header and zero email about it.
That is how most webhook docs updates land. Not with a migration guide. With a quiet edit on one URL.
Webhook pages drift in place
A public webhook reference is not a frozen contract. It drifts:
- the path changes from
/v1/hooksto/v2/webhookswith no redirect - the signature header gets renamed and old events start failing verification ## Waiting for a 404 in production is too late
I used to treat failing deliveries as the monitor. Then I watched a vendor keep the old docs title while the endpoint table swapped under it. We burned half a day chasing our own queue before someone pasted the live docs URL into the thread.
If your process is "we'll notice when events stop," you are noticing after the outage already started for everyone who still posts to the old path.
Watch the exact webhook docs URL
Paste the public webhook, callbacks, or events docs URL your integration README actually links. Ask a watcher for an alert when the page text chang
Start with the endpoints that move money or state
Do not watch every marketing page. Watch the docs that would break a ship:
- webhook / callback references linked from your integration README
- event catalog pages that list payload fields your parser expects
- signature verification sections that rename headers
- "migration" or "changelog" tabs that sometimes stay empty while the main table edits in place
A homepage can still say "reliable webhooks" while /docs/webhooks drops the only path your production worker posts to. Get the alert on that URL, not on the vendor blog.
Originally published on Medium: https://evangelist67.medium.com/how-to-get-a-webhook-docs-alert-before-your-events-start-404ing-4438dc4c721fes, especially around path, signature, header, payload, retry, and version language.
I use AyeWatch for that. Free Preview is $0 with 3 topics and 6 lifetime runs. Pro is $9 a month.
When it fires I open the diff, update the client, and ping the vendor AE the same day. The alert is the triage. The support ticket is optional.
- a required field appears in the payload schema without a version bump
- retry semantics shrink from 24 hours to "best effort"
- the "stable" label stays while the examples below it already point somewhere else
Google Alerts on the product name will not catch a path that leaves one docs URL. Your Slack channel will find out when a customer asks why their sync stopped.
Top comments (0)