AI Summaries.
Every change Ovro detects comes back as one sentence — written by an AI that read the page.
Most change-detection tools hand you a diff and let you figure out the rest. Ovro doesn't. When a watched page changes, an AI vision model reads the new screenshot alongside the previous one and writes you a single summary: 'Price dropped from $349 → $279 (-20%)'. The summary ships with structured fields (old value, new value, delta, severity), a topic tag, and — for high-severity changes — a short explanation of why it matters. The result is a feed that reads like a colleague's morning briefing, not a list of HTML changes.
What's included.
One sentence per change
Every detected change comes back as a single sentence — "Price dropped from $349 → $279 (-20%)" — not a pixel diff or an HTML blob.
Structured before / after
Each summary ships with structured fields: the old value, the new value, a delta count or percentage, and topic tags.
Why it matters
For high-severity changes the AI also writes a short "why this matters" line so you don't have to context-switch to figure out the impact.
Vision-aware
The summary is written from the rendered screenshot, not just the HTML, so it captures changes that only show up after JavaScript runs.
Watch prompts
Tell the AI what you actually care about — "only summarize price changes" — and it filters everything else out before alerting.
Severity classification
Each change is rated low / medium / high so the Insights feed and notifications can prioritize what you see first.
Multi-language
Works across languages. The summary is written in the page's source language so the meaning isn't lost in translation.
Background pipeline
Summaries are generated as a background Inngest job so the change shows up in your feed within seconds of detection.
In detail.
Triggered by every detected change
When Ovro's diff pipeline flags a meaningful change on a watched page, a background Inngest job picks it up, packages the before/after content and screenshots, and hands them to Claude for summarization.
Content extraction first
Before the AI ever sees the page, Ovro uses Mozilla Readability to strip away nav, ads, and footer noise — so the summary describes the actual content change, not a re-render of the chrome.
Structured output
The model is forced to a tool-call schema with explicit fields: label, before, after, delta (text + tone), category, severity, and an optional "why it matters" line. The Insights feed renders each field in its own slot — no string parsing, no regex.
List-mode vs scalar-mode rendering
For multi-item changes (e.g. "three new headlines, two removed"), the UI shows the old list with a strikethrough on top and the new list bold below. For single-value changes (a price flip, a status change) it renders the inline "before → after" pattern with the delta pill.
Severity-aware notifications
The same summary that shows up in the Insights feed is also dispatched to your configured notification channels. Email, Slack, Discord, SMS, or webhooks — they all get the AI line, the visual diff link, and the severity tag.
AI summaries are part of every paid plan. The free tier includes change detection without the AI summary layer.
Try Ovro free.
Free for the basics. Paid plans from $4.99/mo.
Free forever · No credit card