Articles & newsletters
Readable text extracted, clutter dropped, the original always linked. An article you save is the article — not a bookmark that goes stale when the page changes or disappears.
Supported sites
Section titled “Supported sites”| Source | What you get | Chrome extension | Share a link (iOS & Web) |
|---|---|---|---|
| Substack | Full post, clean reading view | ● Optimized | ● Optimized |
| Medium | Full article | ● Optimized | ● Optimized |
| WeChat Official Accounts | Full article, clean reading view | ◐ Works | ● Optimized |
| Wikipedia | Article body with structure | ● Optimized | ● Optimized |
| Any readable web page | Readable text, images, metadata Works even when it is not listed here | ● Optimized | ◐ Works |
● Optimized: tuned for the site · ◐ Works: general capture · ○ Not yet · N/A: not applicable on this path · How capture works
What a save looks like
Section titled “What a save looks like”Every article becomes a Markdown item with:
- The readable body — headings, paragraphs, lists, block quotes, links, and images, in reading order. Navigation, sidebars, cookie banners, and “related posts” are dropped.
- Metadata — title, author when the page declares one, site name, and the original URL.
- A summary and tags, generated on save, so the item is findable before you have read it a second time.
The full text is indexed, so search matches a phrase buried in paragraph twelve, not only the headline.
Which path should I use?
Section titled “Which path should I use?”- Chrome extension — captures the page as you see it, signed in. Use it for subscriber-only newsletters, paywalled articles you have access to, and members-only pages: the extension reads what your browser renders. You get a free preview of the extracted text and summary before saving.
- Share a link — the iOS share sheet, or pasting a URL into the web app. Sensefold visits the page for you as a signed-out visitor. This is the quickest path for public articles and the only path from your phone.
If a page is behind a login and you share the link, Sensefold sees whatever a signed-out visitor sees — usually a teaser or a login wall. Capture those from your browser instead.
Developer sources: GitHub & LeetCode
Section titled “Developer sources: GitHub & LeetCode”Repos and technical pages keep their code blocks intact.
| Source | What you get | Chrome extension | Share a link (iOS & Web) |
|---|---|---|---|
| GitHub | README / issue / discussion content | ● Optimized | ● Optimized |
| LeetCode | Problem statement | ● Optimized | ◐ Works |
● Optimized: tuned for the site · ◐ Works: general capture · ○ Not yet · N/A: not applicable on this path · How capture works
- GitHub — a repository page saves its README; an issue, pull request, or discussion saves the thread with its comments; a file link saves that file’s content (text files only, up to 1 MB). Fenced code blocks stay fenced, so your AI can read a snippet exactly as written.
- LeetCode — the problem statement, with its examples and constraints.
Does it keep code blocks from GitHub issues?
Section titled “Does it keep code blocks from GitHub issues?”Yes. Code fences, inline code, and tables in issues and discussions are kept as Markdown, on both capture paths.
- Any readable public page works through generic extraction even when it is not listed — listed sources have dedicated handling.
- WeChat Official Account articles are best saved through the link path (share from WeChat, or paste the URL), which has tuned extraction.
- Saving is free; the summary and tags spend credits. With zero credits, a link is kept as capture-only — the URL is stored, but the article text is not fetched until you have credits again.
- Everything you save here is readable by your own AI over MCP — see Sensefold for agents — and leaves with you as part of the Markdown ZIP export.