🦅 Hawkwatch
Watch any product page for price or stock changes. Unlimited watchers, your server.
Distill.io meters your checks and caps your watchers at $12–28/month, forever. Hawkwatch points a real headless Chromium at any product page, auto-detects the price from structured data, and pings you — with a full-page screenshot — the moment it drops or restocks. Your server, your schedule, no per-check quotas.
Hawkwatch, as it actually looks — a real screenshot, not a mockup.
Features
Auto price detection
Parses JSON-LD Product/Offer data with og:price and microdata fallbacks — most modern shops work with zero configuration.
CSS selector mode
For pages without structured data, paste a selector. Locale-aware parsing: $1,299.00, 1.299,00 €, 1 299,00 kr all parse correctly.
Stock tracking
From JSON-LD availability, an element selector, or a keyword regex against page text — get pinged the second it flips.
Two fetch engines
Headless Playwright Chromium for JS-rendered shops, or a fast plain HTTP fetch for static pages.
Polite scheduling
Per-watcher interval (5-min floor), ±10% jitter, serialized checks, exponential backoff on failures, robots.txt warnings.
Email + webhook alerts
BYO-SMTP email with the screenshot attached, or raw JSON / Discord / Slack-compatible webhooks. Thresholds: any change, below X, or ±N%.
Price history charts
Every check stored in SQLite — per-watcher line chart with stock band, min/max/current stats, full check log.
Screenshots on change
Full-page capture at every change event, shown in a timeline lightbox and attached to alert emails. Auto-pruned beyond 50 per watcher.
Hawkwatch vs Distill
Distill at $12/mo runs roughly $144/year — $288 over two years. Hawkwatch is $34, once.
| Hawkwatch | Distill | |
|---|---|---|
| Price | $34 once | $12/mo Starter, $28/mo Pro — forever |
| Watchers | Unlimited | Capped per plan |
| Check quotas | None — your server, your schedule | Metered per month |
| Your data | Your SQLite file, on your machine | Their cloud |
| Screenshots on change | Full-page, stored locally | Higher tiers |
| Alerts | BYO SMTP + any webhook | Plan-dependent |
| Self-hosted | Desktop app or any VPS | No |
| Source code | MIT, on GitHub | Closed |
Distill Starter costs $144/year. Hawkwatch pays for itself in under 3 months — and it never bills you again.
Three steps, no subscription
Buy once on Whop
One-time $34 for the packaged Windows installer with updates — Chromium download handled for you.
Run it anywhere
Desktop app with data in your user profile, or Docker on a $5 VPS with Chromium baked into the image.
Add a URL, get alerts
Hawkwatch auto-detects the price, checks on your schedule, and emails or webhooks you — screenshot included — when something moves.
Honest answers
Is it really free on GitHub?
Yes — MIT source at github.com/bensblueprints/hawkwatch, always. The $34 buys the packaged installer, 1-click setup and updates instead of npm and a terminal.
Can I monitor Amazon with it?
Amazon's Terms of Service prohibit automated access, and Hawkwatch won't pretend otherwise — there's no proxy rotation or captcha solving, and there never will be. For Amazon, use their official Product Advertising API (or camelcamelcamel, which is free).
How does it stay polite to the sites I watch?
Checks identify themselves as HawkwatchBot/1.0, intervals floor at 5 minutes with jitter, checks run one at a time with exponential backoff, and adding a watcher warns you (non-blocking) if robots.txt disallows the path.
Do I need a server?
No. Desktop mode runs everything in an Electron window. For 24/7 monitoring while your PC sleeps, a $5 VPS with the included Docker setup is the better home — Distill's cloud does have the always-on part built in, to be fair.
What if a page doesn't have structured data?
Paste a CSS selector and Hawkwatch parses the price out of the element text, locale-aware. Stock can come from a selector or a simple "in stock" keyword match.
Deep-dive comparisons:
Own Hawkwatch forever
$34 once. Deploy on your own server — your data never leaves it. No renewal, no account with us, no meter. Or build it yourself from the MIT source — it's the same app.