HTML to MHTML Converter

Problem

Reviewed pages often lose their layout when remote images break, stylesheets disappear, or sites go offline. Screenshots preserve pixels but not DOM, CSS, or JS; saving raw HTML can leave relative paths and media files scattered. Engineers who need archiving, review, or offline replay need a single self-contained snapshot that can be opened later.

How It Works

The skill takes a required URL and an optional output filename. It fetches the target page and linked resources, then packages HTML, CSS, JavaScript, images, and binary assets into one .mhtml file. The output targets a Chrome-compatible Blink MHTML format, uses quoted-printable for text resources and base64 for images and other binary files, and maintains Content-ID and cid: references so assets can resolve as closely as possible to the original page relationships.

Typical flow:
- Provide an accessible webpage URL;
- Generate or specify an output filename;
- Return the .mhtml file path for opening in Chrome, Edge, or other MHTML-compatible browsers.

Limits

It is better suited to snapshots of static or clearly referenced pages than to restoring full state in complex SPA applications. Pages that depend on login sessions, dynamic scripts, third-party assets, or network policy may be limited by the request environment. Verify resource completeness before relying on the output, and avoid distributing archived pages that contain sensitive information.