hosting
GET /v1/sites/{siteId}/vitals/history
Get a site's speed history and change timeline.
Authentication
Send an API key as a bearer token. This endpoint does not state a specific permission in the specification, so give your key the least it needs and check the response rather than assuming.
This endpoint takes no organisation id. Your key already identifies the organisation it belongs to, and the response is scoped to it.
Try it
Replace anything in angle brackets with your own values, and the key placeholder with a key from your dashboard.
curl -X GET https://api.zinndigital.com/v1/sites/{siteId}/vitals/history \
-H "Authorization: Bearer zdk_live_…"Signed in? The API console in your dashboard fills in your real organisation id and your own key, and runs the request against the live API so you can see the actual response. Open this endpoint in the API console
Details
How the site's speed has moved over time, **and what changed on the site**, on one axis — so a customer can see that activating a plugin cost them 0.4 s rather than having to remember it. **This is the consumer that was missing.** Vitals rows have been kept since #626 with a comment saying they exist "so the Performance tab can show a trend"; nothing ever read them, there was no time-series endpoint, and there was no charting library in the repo. The trend was being stored correctly, for weeks, for nobody. **Two producers, one axis.** `points` are Lighthouse measurements taken off the request path by a scheduled sweep. `events` are what the site reported about itself over a signed webhook (plugin activated, core updated, theme switched, PHP version changed), merged with its successful deploys. Events carry the time they happened **on the site**, so a webhook delivered late still lands beside the score it explains. **`score_delta` is computed here.** Two clients subtracting two nullable columns is two chances to render `-87` for "we did not measure the earlier one". ⚠️ **Cadence.** The sweep measures a fixed number of sites per hour, so on a large estate a site's points can be days apart. A site that reports a change is moved to the front of the next sweep, which is why the timeline — not the sweep — is the primary signal for attributing a regression (docs/85 §4.2). RLS-scoped to a site the caller can view (`sites.view`); an out-of-scope or unknown id is a `404` alike, exactly like `getSite`.
Parameters
| Name | Type | Required | What it is |
|---|---|---|---|
siteId (path) | Uuid | Yes | Site ID (UUIDv7). |
days (query) | integer | No | How far back to look. Clamped rather than rejected — this is a cheap read of the caller's own data, so a `422` on `days=0` would be a worse experience than a sensible window, an… |
Response
| Name | Type | Required | What it is |
|---|---|---|---|
points | SiteVitalsHistoryPoint[] | Yes | Oldest → newest, so a chart renders it without reversing. Capped; when a site has more measurements than the cap, the **most recent** are returned — a chart that stopped before… |
events | SiteEvent[] | Yes | Oldest → newest, on the same axis as `points`. |
stack_type | string | Yes | The site's configured stack — `static`, `php`, `wordpress`, `woocommerce`, `node`, `one_click`. Decides which advice and which events apply: a static site must never be given Wo… |
stack_packs | string[] | Yes | What the newest lab run detected. Resolves `one_click`, which is a container for "whatever the customer installed" and therefore knows nothing about itself. |
opportunities | SiteVitalsOpportunity[] | Yes | The newest lab run's opportunities. On a customer surface these are the same rows `getSiteVitals` returns; on the staff surface they additionally include `owner: platform` audit… |
recommendation | SpeedUpRecommendation | Yes | Which optimisation tier to lead with for this site, and the evidence for it (docs/85 §7). Before this, the speed-up offer fired on a traffic-light band alone while the platform… |
Errors this endpoint can return
401 · 403 · 404 · 429