PixelWatch changelog

PixelWatch now makes monitor state easier to understand and gives the check pipeline stronger rules for retries, failures, and visual comparisons.

• Monitor health is clearer

  The dashboard now distinguishes Healthy, Running, Scheduled, Due, Retrying, Failed, Blocked, and Paused states. Non-healthy states include a reason and the next recommended action, with a debug reference shown for failures when available.

• Scheduled checks handle failures consistently

  Scheduled checks and manual Check Now runs now use the same bounded retry and status handling. Incomplete screenshots or comparisons are not treated as valid diffs or alert events, which keeps partial runs from creating misleading notifications.

• More perceptual comparison options

  The diff engine now supports SSIM, pHash, and dHash comparison modes. Each mode returns a change decision and diff metadata for the comparison pipeline.

• Alert evaluation is more deliberate

  Consecutive-run alert rules now evaluate completed, comparison-valid history before creating an alert. When a rule owns alerting for a monitor, the legacy immediate path is skipped to avoid duplicate alerts for the same match.

The dashboard now makes everyday monitor management easier, from finding a specific site to understanding alerts, setup choices, usage, and plan limits.

• Find monitors faster

  Search monitors by name or URL and filter them by status so you can focus on the sites that need attention.

• Sort the monitor list

  Client-side sorting makes it easier to organize monitored sites without waiting for another page load.

• Export monitored URLs

  Export your monitored URLs to CSV for sharing, review, or use in another workflow.

• Understand alert severity

  The alerts view now includes a severity legend and explanations so the meaning of each alert level is easier to interpret.

• Set up sites with guidance

  The Add Site form validates inputs and provides inline guidance, including help choosing the right page type for the site you want to monitor.

• Understand usage windows

  Usage metrics now explain their time windows so you can read the numbers in the right context.

• See upgrade paths beside gated features

  Upgrade CTAs now appear beside gated features, making the next step clearer when a plan limit applies.

• More consistent feedback states

  Loading, empty, and error states are handled more consistently across the dashboard.

A focused product pass on the monitoring loop: adding sites, running scheduled checks, delivering alerts, and routing feedback reliably.

• Add Site dialog

  Fixed the Add Site flow so the form scrolls and Start Monitoring stays reachable on smaller screens.

• First check on add

  Adding a monitored URL now queues the first screenshot check automatically, so you do not need a manual Check Now to establish the baseline.

• Scheduled checks

  Daily, weekly, and other check intervals now run on a schedule instead of depending only on manual checks.

• Change alert emails

  Visual change alerts email the site owner more reliably, including for scheduled checks that previously could miss the recipient address.

• Admin notifications

  New account signups and new feedback submissions now notify PixelWatch admins by email.

• Feedback to GitHub

  Dashboard feedback creates GitHub issues again, and production fails clearly when GitHub is not configured instead of silently dropping requests.

• Edit monitored sites

  You can now edit an existing monitored URL to update its collection, page type, check interval, diff threshold, and email alert preference.

• Screenshot check capacity

  Screenshot capture now runs on Cloudflare Workers Paid Browser Rendering, which removes the Free-plan daily browser-time bottleneck that stalled checks earlier today.