View Page Source: Inspect HTML, Metadata and Links

Read the source before diagnosing a rendering problem

A page can return successful HTTP source while its visible content depends on a later JavaScript request. This inspector retrieves the source through the SimplyUtils server and extracts its title, named metadata, heading order and anchor destinations. It does not run the inspected website or predict what a search crawler will index.

Original workshop page exercise

Select Load worked sample. This uses a local, deliberately small booking page, with the reference address https://example.com/workshops/booking. It makes no target-site request. The source title is Workshop booking, the description is Book a small workshop., and the heading order is h1 Workshop booking, then h2 Choose a date.

The Contact anchor uses ../contact. The resolved address is https://example.com/contact, because the browser treats booking as a file segment and then moves up from workshops. Help uses //other.example/help, so it inherits HTTPS but has another origin. The email link stays non-HTTP. These three cases demonstrate why classifying every relative-looking link as internal gives misleading results.

On your own page, compare a heading visible on screen with the headings in this source. If it is absent, inspect the rendered DOM in browser developer tools and the live rendered HTML in Search Console. A missing source heading may be inserted by JavaScript; it does not by itself establish an indexing failure. A failed proxy fetch also does not establish that Google cannot access the page.

Resolve links with the right assumptions

The first base element is used if it resolves to an HTTP or HTTPS address without credentials. Otherwise, relative links resolve against your submitted URL. The existing proxy response does not expose the final URL after redirects. If the page redirects into another directory, fetch that final public URL directly before interpreting relative destinations.

Same origin requires matching scheme, domain and port. Protocol-relative links are resolved, fragment anchors are retained, and unsupported schemes remain text. None of the extracted destinations are visited or checked for a successful response. Each metadata, heading and link list is limited to its first 1,000 entries, with a notice when capped. The full accepted source remains available for export.

Fetch, compare and export

Enter a public domain URL; HTTPS is added when no scheme is provided. Credentials, IP literals, private-looking domains and nonstandard ports are rejected by the interface. The server also checks public network destinations and redirects. The browser accepts nonempty HTML up to 1 MiB, while server restrictions may reject a target earlier. JSON endpoints, sign-in pages, bot challenges and unavailable sites cannot be turned into the expected page content here.

Changing the address clears the previous result. Cancel stops the browser request and ignores late results; it cannot guarantee that a request already received by the server stops immediately. A 20-second browser deadline prevents indefinite waiting. Copy reports success only after the clipboard accepts the source. Download exports UTF8 plain text; opening it as text avoids executing remote scripts. If you deliberately rename it to HTML, inspect it before opening it in a browser.

Processing and troubleshooting

Live URLs and returned source pass through the SimplyUtils server, and the target website receives the proxy request. Extraction uses an inert template in your browser; source scripts, images, frames and stylesheets are not attached to the displayed document. Results remain in application memory until the address changes or the page reloads. Consult the Privacy Policy for broader site practices.

If the website looks different from the source, compare the same URL, public access state and redirect destination. This tool provides no HTTP header inspection, sitemap discovery, link health check, social-card rendering or SEO scoring. Use Open Graph Preview for a metadata-based social preview and Search Console for Google's actual rendering test.