We pushed a batch of API improvements on September 14th. Quick rundown of what changed.
Before this, pulling /v1/snapshots for a busy monitor returned everything in insertion order. If you needed a specific date range, you sorted client-side. That was annoying enough that we kept hearing about it. Now all three main list endpoints — monitors, snapshots, diffs — accept sort and order parameters. Snapshots filter by captured_after, captured_before, and status. Monitor queries support url and name search plus tag-based filtering. Diffs take min_change_percent, max_change_percent, and is_significant, so you can ignore minor rendering noise and only surface captures where the page content actually shifted.
Two new endpoints shipped alongside this. GET /v1/monitors/{id}/latest-snapshot returns the most recent screenshot without paging through the full snapshot list — useful if you're displaying current page state in a dashboard. GET /v1/tags returns your tags with monitor counts attached, handy when querying across a large URL portfolio.
Documentation got a full rewrite at the same time. Every endpoint now has code examples in cURL, PHP, Python, and Node.js, with a tab switcher that remembers your preference across sessions. We added four practical guides covering the patterns we see built most often: retrieving a snapshot from a specific date range, detecting whether a page changed between two timestamps, bulk-downloading screenshots, and wiring up webhook alerts. They're the 20–30 line patterns that cover most integration work, not exhaustive tutorials.
Also worth noting: the full OpenAPI spec is at /api/openapi.json for import into Postman, Insomnia, or any compatible client. And llms.txt is live at the site root, following the llms.txt convention — it lists all endpoints and use cases in the format AI coding assistants can read.
API access is on Starter plans and above. Full reference at snapshotarchive.com/docs.
Free plan includes 3 websites with daily captures. No credit card required.
