Save Webpage · Engine

By Matthew Adams 🕢 Published on September 23, 2026 at 7:30 am

↑

Save Webpage Action creates an offline copy of a webpage together with the supporting files it needs to display correctly. It works similarly to a browser's Save As (Complete webpage) feature—Automation Workshop saves the page HTML, downloads referenced resources, and updates the page to use the downloaded local copies.

The result is a portable webpage that can be opened locally without downloading its directly referenced images, stylesheets, scripts, icons, media, and other supported resources again.

From online to offline

Saving a complete webpage involves more than downloading its HTML. Automation Workshop processes the page and its directly referenced resources as a coordinated set.

Retrieve the HTML

Automation Workshop first obtains the webpage HTML from the specified URL, or uses page content supplied by an earlier workflow step.

The HTML remains the foundation of the saved page. Its structure and formatting are preserved while resource references are adjusted where necessary for offline use.

Discover resources

The page is scanned for resources it directly references and needs for display. These may include images, stylesheets, JavaScript files, icons, media, fonts, and embedded content.

Only resources referenced by the webpage itself are discovered. Resources referenced indirectly by a downloaded supporting file are not recursively followed.

Companion folder

When supporting files are found, Automation Workshop creates a companion folder next to the saved HTML file. For example:

• report.html
• report-files\

The companion folder uses the webpage filename with a -files suffix. If that name is already in use, Automation Workshop chooses the next available name, such as report-files (2) or report-files (3).

If the webpage contains no downloadable supporting resources, no companion folder is created and the HTML is saved unchanged.

Reconnect resources

Each discovered resource is downloaded into the companion folder. Automation Workshop then rewrites the corresponding references in the saved HTML so they point to the local files instead of their original online locations.

The final HTML file and its companion folder form a single offline webpage copy.

What is saved?

Save Webpage engine focuses on resources used to render and present the page…

Images and visuals

Downloaded image resources can include:

• Regular webpage images.
• Responsive image alternatives and multiple resolutions.
• <picture> sources.
• Video poster images.
• Legacy background images used by pages, tables, or cells.
• Images referenced directly by inline styles or <style> blocks.
• SVG artwork directly referenced by the page, including sprite files while preserving fragment references such as sprite.svg#star.

Stylesheets and scripts

External CSS stylesheets referenced by the page are downloaded, as are JavaScript files referenced through page markup.

Automation Workshop also recognizes resources referenced directly from CSS contained within the HTML itself, including @import stylesheets, fonts, and background, border, or cursor images referenced by inline styles and <style> blocks.

Icons and app resources

Supported page resources also include:

• Favicons.
• Touch icons.
• Mask icons.
• Web app manifests.

Media and embedded content

The engine can also retrieve:

• Video files.
• Audio files.
• Subtitle and caption tracks.
• Content referenced by <embed> and <object> elements.

These categories represent resources directly referenced by the page and used as part of its displayed content.

Subresource handling

Problems encountered while retrieving subresources are treated as warnings rather than failures of the entire Action.

For example, if the primary HTML resource is retrieved successfully with a 200 OK response but one or more referenced images, stylesheets, scripts, or other assets are unavailable, Save Webpage can still complete successfully. The resulting archive contains the primary resource together with all subresources that could be retrieved, providing the most complete offline copy possible.

When the KeepOriCompression extra setting is enabled, the primary response body is preserved in its original compressed form.

Because Automation Workshop does not decompress or inspect that content, referenced subresources cannot be discovered or processed. This behavior is inherent to preserving the original compressed response unchanged.

What is not saved?

Save Webpage intentionally does not turn the entire linked website into an offline mirror. Certain references remain unchanged.

Links and targets

Normal navigation links such as <a href> are not downloaded or rewritten into local page copies. This also applies to:

• Image-map destinations.
• Form submission targets.
• Links that happen to point directly to PDFs, images, or other downloadable files.

A linked file is not considered a page resource merely because it can be downloaded.

Frames and embedded docs

Content loaded through <iframe> elements or inline frame documents is not followed or saved.

Metadata and hints

Resources used only for metadata or browser hints are left unchanged, including:

• Open Graph and social preview images.
• Twitter preview metadata.
• Page-refresh metadata.
• Canonical links.
• preconnect and dns-prefetch hints.
• Author and license links.
• Alternate-language links.

Inline and hidden references

Content already embedded directly in the HTML through data: URLs remains in place because it is already self-contained.

URLs that appear only inside HTML comments or JavaScript text are not treated as downloadable webpage resources.

Direct references only

A key distinction is that Save Webpage discovers resources referenced directly by the webpage, not an unlimited chain of dependencies.

For example, if the HTML references site.css, that stylesheet itself is downloaded. However, if site.css references another stylesheet, a font, or a background image, those nested resources are not discovered automatically.

Resources referenced directly by inline CSS or <style> blocks in the webpage are supported because those references are part of the page itself.

Names and storage

Automation Workshop derives local filenames from the resource addresses whenever possible.

Best-effort naming

File extensions and names are determined on a best-effort basis. When the resource type cannot be reliably determined from its web address—for example, an image delivered from an address ending in pic.php—the resulting local filename may have an imperfect or missing extension, which cannot always be corrected after the file has been downloaded.

Names are also adjusted when necessary to make them valid Windows filenames. Invalid characters are removed, reserved Windows names are handled safely, and excessively long names are shortened.

Duplicate names

Different web resources are never allowed to overwrite one another simply because their local names would be identical. For example:

• logo.png
• logo (2).png
• logo (3).png

The saved HTML is updated to reference the correct local copy.

Repeated references

If the same resource address appears multiple times in the page, Automation Workshop downloads it only once. All repeated references share that single local file.

Offline compatibility

Some security information that is valid for an online resource can prevent that same resource from loading correctly from a local file.

To make the saved page usable offline, Automation Workshop removes selected security-related attributes that bind resources to their original web location, including integrity and cross-origin checks where applicable. A page-level Content Security Policy meta tag is also removed from the saved copy when necessary.

The saved webpage is therefore intended as an offline representation of the page, not as an exact replica of the original website's security environment. To retrieve web content without these offline compatibility adjustments, use Get Webpage Content or Download File from Website Actions instead.

Save Webpage supports UTF-8 webpages. Pages using other character encodings are outside the supported scope of this feature.

Preserved page structure

Save Webpage does not rebuild or redesign the webpage. The original HTML structure, content, and formatting are retained as received. Only references that need to point to downloaded local resources are changed.

This helps the offline copy remain as close as practical to the webpage returned by the server while keeping the saved resources together in a predictable local structure.

Summary

Save Webpage produces two closely connected parts when supporting resources are present:

• The saved HTML page.
• A companion -files folder containing its downloaded resources.

Together they provide a convenient offline copy while avoiding unrelated navigation pages, frames, metadata-only resources, and recursively referenced files that are not directly part of the original webpage.

Just ask…

If you have any questions, please do not hesitate to contact our support team.