Change Detection

Try our Chrome extension

Easily add the current web-page from your browser directly into your changedetection.io tool, more great features coming soon!

Changedetection.io needs your support!

You can help us by supporting changedetection.io on these platforms;

The more popular changedetection.io is, the more time we can dedicate to adding amazing features!

Many thanks :)

changedetection.io team

General
Notifications
Fetching
Global Filters
UI Options
API
RSS
Backups
Time & Date
CAPTCHA & Proxies
AI / LLM
Info

Time Between Check

Weeks
Days
Hours
Minutes
Seconds

Default recheck time for all watches, current system minimum is 3 seconds (more info).

Use time scheduler

Set a hourly/week day schedule

Schedule time limitsBusiness hours Weekends Reset

Monday
Start At
Run duration
Hours
Minutes
Tuesday
Start At
Run duration
Hours
Minutes
Wednesday
Start At
Run duration
Hours
Minutes
Thursday
Start At
Run duration
Hours
Minutes
Friday
Start At
Run duration
Hours
Minutes
Saturday
Start At
Run duration
Hours
Minutes
Sunday
Start At
Run duration
Hours
Minutes
Warning, one or more of your 'days' has a duration that would extend into the next day.
This could have unintended consequences.
Optional timezone to run in

More help and examples about using the scheduler

Number of times the filter can be missing before sending a notification

After this many consecutive times that the CSS/xPath filter is missing, send a notification
Set to 0 to disable

Number of history items per watch to keep

Limit collection of history snapshots for each watch to this number of history items.
Set to empty to disable / no limit

Password

Password protection for your changedetection.io application.

Allow anonymous access to watch history page when password is enabled

Allow access to the watch change history page when password is enabled (Good for sharing the diff page)

Treat empty pages as a change?

When a request returns no content, or the HTML does not contain any text, is this considered a change?

Notification URL List

Tip: Use AppRise Notification URLs for notification to just about any service! Please read the notification services wiki here for important configuration notes.

Show advanced help and tips

Send test notification Notification debug logs

Notification Title

Title for all notifications

Notification Body

Body for all notifications — You can use Jinja2 templating in the notification title, body and URL, and tokens from below.

Show token/placeholders

Token	Description
`{{base_url}}`	The URL of the changedetection.io instance you are running.
`{{watch_url}}`	The URL being watched.
`{{watch_uuid}}`	The UUID of the watch.
`{{watch_title}}`	The page title of the watch, uses <title> if not set, falls back to URL
`{{watch_tag}}`	The watch group / tag
`{{preview_url}}`	The URL of the preview page generated by changedetection.io.
`{{change_datetime}}`	Date/time of the change, accepts format=, change_datetime(format='%A')', default is '%Y-%m-%d %H:%M:%S %Z'
`{{diff_url}}`	The URL of the diff output for the watch.
`{{diff}}`	The diff output - only changes, additions, and removals All diff variants accept `lines=`, `context=`, `word_diff=`, `ignore_junk=` args, e.g. `{{diff(lines=10)}}`, `{{diff_added(lines=5, context=2)}}`
`{{diff_clean}}`	The diff output - only changes, additions, and removals — Without (added) prefix or colors
`{{diff_added}}`	The diff output - only changes and additions
`{{diff_added_clean}}`	The diff output - only changes and additions — Without (added) prefix or colors
`{{diff_removed}}`	The diff output - only changes and removals
`{{diff_removed_clean}}`	The diff output - only changes and removals — Without (added) prefix or colors
`{{diff_full}}`	The diff output - full difference output
`{{diff_full_clean}}`	The diff output - full difference output — Without (added) prefix or colors
`{{diff_patch}}`	The diff output - patch in unified format
`{{diff_changed_from}}`	Only the changed words/values from the previous version — e.g. the old price. Best when a single value changes per line; multiple changed fragments are joined by newline.
`{{diff_changed_to}}`	Only the changed words/values from the new version — e.g. the new price. Best when a single value changes per line; multiple changed fragments are joined by newline.
`{{current_snapshot}}`	The current snapshot text contents value, useful when combined with JSON or CSS filters
`{{triggered_text}}`	Text that tripped the trigger from filters
`{{diff}}` (upgraded)	When AI Change Summary is configured, contains the AI-generated description instead of the raw diff. Falls back to raw diff when not configured.
`{{raw_diff}}`	Always the raw +/- diff, regardless of AI Change Summary setting.
`{{llm_summary}}`	The AI Change Summary text (same as the upgraded {{diff}} — explicit reference).
`{{llm_intent}}`	The AI Change Intent that was evaluated.

Warning: Contents of {{diff}}, {{diff_removed}}, and {{diff_added}} depend on how the difference algorithm perceives the change.
For example, an addition or removal could be perceived as a change in some cases. More Here

For JSON payloads, use |tojson without quotes for automatic escaping, for example - { "name": {{ watch_title|tojson }} }
URL encoding, use |urlencode, for example - gets://hook-website.com/test.php?title={{ watch_title|urlencode }}
Regular-expression replace, use |regex_replace, for example - {{ "hello world 123" | regex_replace('[0-9]+', 'no-more-numbers') }}
For a complete reference of all Jinja2 built-in filters, users can refer to the https://jinja.palletsprojects.com/en/3.1.x/templates/#builtin-filters

Notification format

Format for all notifications

Discord does not render HTML — switch to Plain Text format to avoid   and other HTML entities appearing literally in your notifications.

Notification base URL override

Base URL used for the {{ base_url }} token in notification links.
Default value is the system environment variable 'BASE_URL' - read more here.

Fetch Method

Basic fast Plaintext/HTTP Client
WebDriver Chrome/Javascript

Use the Basic method (default) where your watched sites don't need Javascript to render.

The Chrome/Javascript method requires a network connection to a running WebDriver+Chrome server, set by the ENV var 'WEBDRIVER_URL'.

If you're having trouble waiting for the page to be fully rendered (text missing etc), try increasing the 'wait' time here.
This will wait n seconds before extracting the text.

Wait seconds before extracting text

Number of fetch workers

Number of concurrent workers to process watches. More workers = faster processing but higher memory usage.
Currently running: 5 operational async workers.

Random jitter seconds ± check

Example - 3 seconds random jitter could trigger up to 3 seconds earlier or up to 3 seconds later

Requests timeout in seconds

For regular plain requests (not chrome based), maximum number of seconds until timeout, 1-999.

Default User-Agent overrides

Plaintext requests

Applied to all requests.

Note: Simply changing the User-Agent often does not defeat anti-robot technologies, it's important to consider all of the ways that the browser is detected.

Tip: Connect using Bright Data proxies, find out more here.

Ignore whitespace

Ignore whitespace, tabs and new-lines/line-feeds when considering if a change was detected.
Note: Changing this will change the status of your existing watches, possibly trigger alerts etc.

Render anchor tag content

Render anchor tag content, default disabled, when enabled renders links as (link text)[https://somesite.com]
Note: Changing this could affect the content of your existing watches, possibly trigger alerts etc.

Remove elements

Remove HTML element(s) by CSS and XPath selectors before text conversion.
Don't paste HTML here, use only CSS and XPath selectors
Add multiple elements, CSS or XPath selectors per line to ignore multiple parts of the HTML.

Ignore Text

Note: This is applied globally in addition to the per-watch rules.

Matching text will be ignored in the text snapshot (you can still see it but it wont trigger a change)
Note: This is applied globally in addition to the per-watch rules.
Each line processed separately, any line matching will be ignored (removed before creating the checksum)
Regular Expression support, wrap the entire line in forward slash /regex/
Changing this will affect the comparison checksum which may trigger an alert

Strip ignored lines

Remove any text that appears in the "Ignore text" from the output (otherwise its just ignored for change-detection)
Note: Changing this will change the status of your existing watches, possibly trigger alerts etc.

API Access

Drive your changedetection.io via API, More about API access and examples here.

API access token security check enabled

Restrict API access limit by using x-api-key header - required for the Chrome Extension to work

API Key 893ff0c397d7dd1c84c1b9a055ffeca1

Regenerate API key

Chrome Extension

Easily add any web-page to your changedetection.io installation from within Chrome.

Step 1 Install the extension, Step 2 Navigate to this page, Step 3 Open the extension from the toolbar and click "Sync API Access"

Chrome Webstore

Hide muted watches from RSS feed

Number of changes to show in watch RSS feed

Maximum number of history snapshots to include in the watch specific RSS feed.

Enable RSS reader mode

For watching other RSS feeds - When watching RSS/Atom feeds, convert them into clean text for better change detection.

RSS Content format

Does your reader support HTML? Set it here

RSS body built from

'System default' for the same template for all items, or re-use your "Notification Body" as the template.

RSS "System default" template override

Body for all notifications — You can use Jinja2 templating in the notification title, body and URL, and tokens from below.

Show token/placeholders

Token	Description
`{{base_url}}`	The URL of the changedetection.io instance you are running.
`{{watch_url}}`	The URL being watched.
`{{watch_uuid}}`	The UUID of the watch.
`{{watch_title}}`	The page title of the watch, uses <title> if not set, falls back to URL
`{{watch_tag}}`	The watch group / tag
`{{preview_url}}`	The URL of the preview page generated by changedetection.io.
`{{change_datetime}}`	Date/time of the change, accepts format=, change_datetime(format='%A')', default is '%Y-%m-%d %H:%M:%S %Z'
`{{diff_url}}`	The URL of the diff output for the watch.
`{{diff}}`	The diff output - only changes, additions, and removals All diff variants accept `lines=`, `context=`, `word_diff=`, `ignore_junk=` args, e.g. `{{diff(lines=10)}}`, `{{diff_added(lines=5, context=2)}}`
`{{diff_clean}}`	The diff output - only changes, additions, and removals — Without (added) prefix or colors
`{{diff_added}}`	The diff output - only changes and additions
`{{diff_added_clean}}`	The diff output - only changes and additions — Without (added) prefix or colors
`{{diff_removed}}`	The diff output - only changes and removals
`{{diff_removed_clean}}`	The diff output - only changes and removals — Without (added) prefix or colors
`{{diff_full}}`	The diff output - full difference output
`{{diff_full_clean}}`	The diff output - full difference output — Without (added) prefix or colors
`{{diff_patch}}`	The diff output - patch in unified format
`{{diff_changed_from}}`	Only the changed words/values from the previous version — e.g. the old price. Best when a single value changes per line; multiple changed fragments are joined by newline.
`{{diff_changed_to}}`	Only the changed words/values from the new version — e.g. the new price. Best when a single value changes per line; multiple changed fragments are joined by newline.
`{{current_snapshot}}`	The current snapshot text contents value, useful when combined with JSON or CSS filters
`{{triggered_text}}`	Text that tripped the trigger from filters
`{{diff}}` (upgraded)	When AI Change Summary is configured, contains the AI-generated description instead of the raw diff. Falls back to raw diff when not configured.
`{{raw_diff}}`	Always the raw +/- diff, regardless of AI Change Summary setting.
`{{llm_summary}}`	The AI Change Summary text (same as the upgraded {{diff}} — explicit reference).
`{{llm_intent}}`	The AI Change Intent that was evaluated.

Ensure the settings below are correct, they are used to manage the time schedule for checking your web page watches.

UTC Time & Date from Server: 2026-05-06T21:28:30.391017+00:00

Local Time & Date in Browser:

Default timezone for watch check scheduler

Open 'History' page in a new tab

Enable this setting to open the diff page in a new tab. If disabled, the diff page will open in the current tab.

Realtime UI Updates Enabled

Realtime UI Updates Enabled - (Restart required if this is changed)

Favicons Enabled

Enable or Disable Favicons next to the watch list

Use page <title> in watch overview list

Pager size

Number of items per page in the watch overview list, 0 to disable.

BrightData offer world-class proxy services, "Data Center" proxies are a very affordable way to proxy your requests, whilst WebUnlocker can help solve most CAPTCHAs.

BrightData offer many many different types of proxies, it is worth reading about what is best for your use-case.

When you have registered, enabled the required services, visit the API example page, then select Python, set the country you wish to use, then copy+paste the access Proxy URL into the "Extra Proxies" boxes below.

The Proxy URL with BrightData should start with http://brd-customer...

When you sign up using https://brightdata.grsm.io/n0r16zf7eivq BrightData will match any first deposit up to $150

Collect public data at scale with industry-leading web scraping solutions and the world’s largest ethical proxy network.

Oxylabs also provide a WebUnlocker proxy that bypasses sophisticated anti-bot systems, so you don’t have to.

Serve over 195 countries, providing Residential, Mobile and ISP proxies and much more.

Use the promo code boost35 with this link https://oxylabs.go2cloud.org/SH2d for 35% off Residential, Mobile proxies, Web Unblocker, and Scraper APIs. Built-in proxies enable you to access data from all around the world and help overcome anti-bot solutions.

Tip: "Residential" and "Mobile" proxy type can be more successful than "Data Center" for blocked websites.

Extra Proxies

Extra Proxies-0

Name
Proxy URL

Extra Proxies-1

Name
Proxy URL

Extra Proxies-2

Name
Proxy URL

Extra Proxies-3

Name
Proxy URL

Extra Proxies-4

Name
Proxy URL

"Name" will be used for selecting the proxy in the Watch Edit settings
SOCKS5 proxies with authentication are only supported with 'plain requests' fetcher, for other fetchers you should whitelist the IP access instead

Extra Browsers can be attached to further defeat CAPTCHA's on websites that are particularly hard to scrape.
Simply paste the connection address into the box, More instructions and examples here

Extra Browsers

Extra Browsers-0

Name
Browser connection URL

Extra Browsers-1

Name
Browser connection URL

Extra Browsers-2

Name
Browser connection URL

Extra Browsers-3

Name
Browser connection URL

Extra Browsers-4

Name
Browser connection URL

✦ AI-powered change monitoring

Connect an LLM to move from "something changed" to "only the thing you care about changed".

⊞

Intent filtering

Each watch or tag can carry a plain-text intent — "notify me only when the price drops" or "alert when the item goes out of stock". On every detected change the AI evaluates the diff against it and suppresses irrelevant noise.

≡

AI Change Summary

Instead of raw diffs, receive plain-language summaries in notifications — "Price dropped from $89 to $67" or "3 new items added to the listing". Set a global default prompt here, or override per watch or tag.

≈

Minimal cost

The AI sees only a unified diff of what changed — never full page HTML. Low-cost models like gpt-4o-mini or Gemini Flash handle this well, typically fractions of a cent per check.

✓ AI / LLM configured: gemini/gemini-pro-latest

AI Provider

Provider

API Key

API Base URL

Only needed for Ollama or custom/self-hosted endpoints. Leave blank for cloud providers.

Available models

Model

✓ AI / LLM configured: gemini/gemini-pro-latest ✕ Remove

Your API key is stored locally and sent only to your chosen provider. On each detected change, the watch's diff and extracted text are sent to the LLM — no full page HTML.

Cache ✕ Clear all summary cache Removes all cached AI change summaries across all watches. They will be regenerated on the next check.

Default AI Change Summary

Default AI Change Summary prompt

Used for all watches unless overridden by the watch or its tag/group. Modify default prompt

Behaviour

Replace {{diff}} notification token with AI summary When enabled, the {{diff}} notification token shows the AI summary instead of the raw diff. Use {{raw_diff}} to always get the original.

Use LLM as a fallback for extracting price and restock info When enabled, the AI will be used as a last resort to extract price and stock status from product pages where no structured metadata (JSON-LD, microdata, OpenGraph) is found.

AI thinking budget (tokens) For Gemini 2.5+ models only. Thinking tokens improve reasoning quality but count against the output budget. Set to Off if summaries are being cut short.

Max AI summary length (tokens) Upper limit on tokens the AI may use when writing a change summary. Higher values allow longer summaries but cost more.

When monthly token budget is reached

Skip AI summarisation only (watch still checks) Skip the watch check entirely

Token & Cost Tracking

No AI usage recorded yet.

Token budget tokens per month (0 = unlimited)

Max input characters characters — currently enforcing: 100,000

Uptime: 1 hour, 2 minutes, 21 seconds

Python version: 3.11.15

Plugins active:

levenshtein_plugin - Levenshtein distance and similarity plugin for text change detection.
wordcount_plugin - Word count plugin for content analysis.
llm_restock - LLM fallback plugin for price and restock info extraction.

Back Clear Snapshot History