Website Change Detector

Schedule this Actor on Apify (or trigger it from n8n / Make) to watch public web pages and get a structured change report whenever the text moves.

The first run stores a baseline snapshot per URL and is labelled baseline — later runs compare with sklearn TF-IDF cosine (changeScore) plus a human-readable difflib diff. Pay per result, not a $15/month rental.

Use it for price pages, terms of service, job boards, and competitor landing pages. HTTP + BeautifulSoup (no Playwright unless a page is JS-only).

What does Website Change Detector do?

It fetches each URL with HTTP, extracts visible text (or cleaned HTML), hashes it, and compares it to the previous snapshot in a named key-value store. Optional CSS selectors focus on a region; ignore selectors drop cookie banners and timestamps.

It does not screenshot pages by default and does not log in to private apps.

Why monitor website changes?

• Competitor price and packaging pages
• Policy / ToS compliance diffs
• Career-page job-list updates
• News homepage alerts

Apify Schedule plus n8n/Make webhooks turn a diff into a Slack message.

What data can it extract?

Field	Type	Description
url	string	Checked URL
changeStatus	string	baseline, unchanged, changed, error
changeScore	number	1 − TF-IDF cosine vs previous text
diff	string	Unified diff
addedChars / removedChars	number	Size of the edit
contentHash	string	SHA-256 of compared content

How to detect website changes

1. Try the prefilled https://apify.com and https://github.com example with mode: text.
2. The first run is the baseline. Schedule hourly/daily afterwards.
3. Filter dataset rows where changeStatus is changed.
4. Pipe diff into Slack or a ticket.
5. Call technicaldost/website-change-detector from MCP.

How much does it cost?

Pay per event. Actor start is $0.00005 (per GB of memory). Each result row is $0.005. This is not a monthly rental. Confirm the live Pricing tab before a large URL list.

Input

See the input tab. Example:

{
  "urls": ["https://apify.com", "https://github.com"],
  "mode": "text"
}

Output

Download JSON, CSV, HTML, or Excel.

{
  "url": "https://apify.com",
  "changeStatus": "baseline",
  "changeScore": 0,
  "title": "Apify: Full-stack web scraping and data extraction platform"
}

MCP

Apify MCP server, actor technicaldost/website-change-detector, same JSON.

FAQ and support

Only public pages. Honour robots.txt and site terms. Open Issues for bugs.