What Is a Broken Image Finder?
Quick answer: A broken image finder scans a web page, extracts every image URL and requests each one to see whether it loads. The ImageFix Broken Image Finder reports broken URLs, soft 404 images, missing width and height attributes, oversized files and formats that browsers cannot display, so you can fix them before visitors and search engines notice.
Key takeaways
- A broken image is any image URL that returns an error, an empty file, or an HTML page instead of picture data.
- Missing width and height attributes make the page jump while it loads and hurt Cumulative Layout Shift.
- Oversized images slow down loading and often delay the Largest Contentful Paint.
- Unsupported formats such as HEIC, TIFF and PSD show as empty boxes in most browsers.
- The scanner is built with SSRF protection, so it can never reach private or internal addresses.
What the Scanner Checks
| Check | What it finds | Why it matters |
|---|---|---|
| Broken URLs | HTTP 404, 410 and 5xx errors, DNS and SSL failures, timeouts, empty files, HTML returned as an image | Visitors see empty boxes, and crawlers cannot index the picture |
| Missing dimensions | img tags without a width or a height attribute | The browser cannot reserve space, so the layout shifts |
| Oversized assets | Heavy files, very wide images, and images delivered much larger than they display | Slow pages, wasted bandwidth, poor Core Web Vitals |
| Unsupported formats | HEIC, HEIF, TIFF, PSD, camera RAW, JPEG 2000, JPEG XL, BMP (detected from the file bytes) | Browsers cannot show the image, or they show it with poor efficiency |
| Redirects and slow responses | Image URLs that hop through redirects or answer slowly | Extra requests delay rendering |
| Mixed content | http:// images on https:// pages | Browsers may block or upgrade them |
| Missing alt text | img tags without an alt attribute | Accessibility and image search context |
Why Broken and Heavy Images Hurt Your Site
Visitors judge a page in seconds. An empty image box looks like a bug, and a slow hero image makes people leave. Search engines also read image URLs, alt text and page speed signals, so broken images weaken a page that otherwise ranks well. Google’s image best practices ask you to use standard HTML image elements, descriptive alt text and fast, supported formats.
Images break for predictable reasons: a file moved during a redesign, a typo in a path, a CDN that expired a signed URL, a hotlink rule that blocks outside requests, a case-sensitive server that treats Photo.JPG and photo.jpg as different files, or an editor who pasted a link to a file on their own computer.
How to Find Broken Images on a Website
- Enter your page URL. Type or paste the public address of the page you want to audit, for example your blog post or product page.
- Choose the scan scope. Scan one page, or scan up to ten pages that it links to on the same site. Pick which image sources to include.
- Run the scan. Click Scan images. The tool fetches the page, extracts the images and checks each one while a progress bar shows the status.
- Review the issues. Filter by broken URLs, missing dimensions, oversized assets, unsupported formats, redirects, missing alt text and mixed content. Open a row for details.
- Fix and export. Fix the issues in your CMS or code, rescan to confirm, and export the CSV report for your team or client.
How to Fix Each Image Problem
| Problem | Fix |
|---|---|
| 404 or 410 image | Restore the file, correct the path, or replace the image, and add a redirect when you move files on purpose. |
| HTML returned instead of an image (soft 404) | Fix the URL and make the server return a real 404 status for missing files. |
| HTTP 403 or 429 for the scanner | Check hotlink and bot rules. Allow the ImageFixBot user agent or test the URL in a browser. |
| Missing width and height | Add both attributes with the file’s real pixel size and let CSS scale the image, for example max-width:100%;height:auto. |
| Oversized file or dimensions | Resize to the largest size you display, compress it, and serve WebP or AVIF with a JPG fallback. Use srcset for responsive sizes. |
| Unsupported format | Convert the file to JPG, PNG, WebP or AVIF before you publish it. |
| Mixed content | Change the image URL to https:// and make sure the host serves a valid certificate. |
| Redirected image URL | Update the HTML to the final URL so the browser skips the extra hop. |
After you fix the files, shrink them with the Image Compressor, resize them with the Image Resizer and convert unsupported files with the All-in-One Image Converter.
How the Scanner Stays Safe: SSRF Protection
A tool that fetches URLs on your behalf can attack internal networks if it trusts every address. This scanner blocks that risk with layered controls that follow the OWASP SSRF prevention guidance:
- Strict URL rules. Only http and https on ports 80 and 443 pass, and URLs with embedded credentials fail.
- Own DNS resolution. The server resolves the hostname itself and blocks private, loopback, link-local, carrier-grade NAT, multicast, documentation, NAT64, 6to4 and IPv4-mapped addresses. If any returned address is blocked, the whole host is blocked.
- Address pinning. The connection uses the address that passed the check, which stops DNS rebinding.
- Redirect re-validation. The scanner follows redirects manually, up to five hops, and validates every hop again.
- Allow-list for images. The server only requests image URLs that it extracted from the pages you scanned.
- Resource limits. Pages are capped at 2 MB, images are read in small ranges, every request has a timeout, and each network address has a request budget.
- Polite crawling. The scanner identifies itself as ImageFixBot and follows robots.txt rules for pages.
What the Scanner Cannot See
The scanner reads raw HTML and does not run JavaScript. Images that a script builds after load, images that need a login, and pictures referenced only inside external stylesheets do not appear in the report. It checks up to 200 unique images and up to ten pages per scan. A refused request (HTTP 403 or 429) means the server blocked the scanner, not that the image is missing, so the report labels it separately. For thorough audits of dynamic sites, combine this report with a browser-based test.
Frequently Asked Questions
What is a broken image finder?
A broken image finder scans a web page, collects every image URL it finds, and requests each one to see whether it loads. It reports images that return errors, images that are not really images, and images that are too big or in a format browsers cannot show.
How does the Broken Image Finder work?
You enter a public URL. The server fetches the page, reads the HTML, extracts the image URLs, and then requests each image with a small range request to read its status, size, format and dimensions. The tool does not run JavaScript, so it sees the same images a search engine crawler sees in the raw HTML.
How does the tool prevent SSRF attacks?
The scanner accepts only http and https URLs on ports 80 and 443, refuses credentials in URLs, resolves the hostname itself and blocks private, loopback, link-local, carrier-grade NAT and other reserved addresses. It pins the checked address for the connection, validates every redirect again, limits response size and time, and only fetches image URLs that it extracted from the scanned pages.
Why does the tool report an image as broken when it loads in my browser?
Some servers refuse requests from bots with HTTP 403 or 429, or they use hotlink protection, geo rules or cookies. The tool marks these as refused by the scanner rather than missing. Images that JavaScript builds after the page loads are not visible to the scanner at all.
What counts as a missing dimension?
An img tag without both a width and a height attribute counts as missing dimensions. Without them the browser cannot reserve space before the image loads, so the page jumps while it loads. This hurts the Cumulative Layout Shift metric.
What counts as an oversized image?
By default the tool flags files above 500 KB, images wider than 2400 pixels, and images whose natural width is more than twice the width set in the HTML. You can change all three limits in the settings and the report updates instantly.
Which image formats does the tool call unsupported?
The tool flags HEIC, HEIF, TIFF, PSD, camera RAW files and Windows metafiles because browsers cannot display them. It warns about JPEG 2000, JPEG XL and BMP because they have limited support or waste bytes. It detects the real format from the file bytes, not just from the extension.
Does the scanner crawl my whole website?
No. It scans one page or up to ten pages that the start page links to on the same host, and it checks up to 200 unique images per scan. It reads robots.txt before it fetches a page and identifies itself as ImageFixBot.
Can the tool scan pages behind a login?
No. The scanner only reads publicly accessible pages and sends no cookies. Scan a public staging copy, or export the page HTML and check the images in your own environment.
Does the tool store my URL or my results?
The results stay in your browser. The server keeps the list of image URLs it found in your session for up to one hour so that it can verify later requests, and it keeps short-lived request counters tied to a hashed network address to limit abuse. It does not keep a scan history.
Related ImageFix Tools
- Image CompressorCut file size without visible loss
- Image ResizerScale images to the displayed size
- All-in-One Image ConverterConvert HEIC, TIFF and more to web formats
- Transparent Image Edge CheckerFind halos and stray pixels in PNG cutouts
- EXIF Location ScrubberRemove GPS data before you publish photos
- All image toolsBrowse the full ImageFix toolkit