Site Health Watch – Daily SEO Accident Alerts

Find out the morning after – not three weeks later in Search Console – when your site starts disappearing from Google.

Schedule this Actor daily. It reads your sitemap, checks every page, compares with the previous run and puts new accidents first:

Severity	Accident
🔴 critical	robots.txt blocks the whole site for Googlebot, Bingbot or all crawlers (classic: staging robots.txt deployed to production)
🔴 critical	noindex on a page that is in your sitemap (meta robots, googlebot meta or X-Robots-Tag header)
🔴 critical	Sitemap URL returns 404 / 410 / 500
🔴 critical	Sitemap can't be found or read
🟡 warning	Canonical points to another page (e.g. every page canonicalised to the home page after a template change)
🟡 warning	Sitemap URL redirects, page has no <title>, page could not be checked
⚪ info	URLs added to / removed from the sitemap, robots.txt edited, titles changed

On a quiet day you get one summary row: ok – No problems found on 184 page(s).

Why this one

Most link / status / SEO checkers on the Store run once and list everything every time. Site Health Watch remembers the last run (in a named key-value store in your account) and tells apart:

• new problems – marked isNew, sorted first, and counted in the summary (2 critical NEW since last run)
• problems that were already there – still listed (or hidden with Only report new problems and changes)
• problems that were fixed since last time – counted in the summary

So you can wire it to an email / Slack integration on the summary status new-critical and only be pinged when something actually broke.

Polite by design

• Only the sites you list. It reads robots.txt first and never fetches a disallowed URL; if robots.txt returns a server error it stops (RFC 9309).
• One request at a time, at least 1 second apart (default 2 s), longer if Crawl-delay says so.
• Identifies itself: SiteHealthWatch/1.0. Collects no personal data.

Input

{
  "siteUrls": [{ "url": "https://www.example.com/" }],
  "maxPagesPerSite": 500,
  "onlyNewAndChanged": true
}

Then Schedules → Create schedule → daily. The first run saves a baseline (current problems are listed, none marked new).

Output

[
  { "type": "summary", "site": "https://www.example.com", "status": "new-critical",
    "message": "3 critical, 1 warning(s) – 2 critical NEW since last run. 1 earlier problem(s) fixed.",
    "critical": 3, "warnings": 1, "newIssues": 2, "fixed": 1, "pagesChecked": 184, "sitemapUrls": 184 },
  { "type": "page-noindex", "severity": "critical", "isNew": true, "url": "https://www.example.com/pricing",
    "message": "noindex is set on a URL listed in the sitemap – it will be removed from search results" },
  { "type": "urls-removed", "severity": "info", "message": "2 URL(s) removed from the sitemap", "urls": ["https://www.example.com/old-a", "https://www.example.com/old-b"] }
]

Pricing

Pay per event, platform usage included:

Event	Price
Actor start	$0.01 per run
page-checked – one page checked	$0.50 per 1,000 pages

Example: a 200-page site checked every day ≈ 30 × ($0.01 + 200 × $0.0005) = $3.30 per month. Pages skipped because of robots.txt or errors are free. A run without a site only returns a setup reminder.

日本語での説明

自分のサイトを毎日チェックし、「検索から消える事故」が起きたときだけ知らせます。たとえば、robots.txt でサイト全体をブロックしてしまった、sitemap に載っているページに noindex が付いた、404・500 になった、canonical が別のページを指すようになった、などです。前回の結果を覚えているので、新しく起きた問題を先頭に出し、直った問題も数えます。robots.txt を守り、1秒以上の間隔を空けてアクセスします。

料金: 1回 $0.01 ＋ 1ページ $0.0005（200ページのサイトを毎日見ると、月 約 $3.30）。