SMF·INSTAPAPER-PREMIUM
Can a prompt replace Instapaper Premium?
Read it later & RSS — read-later, bookmarks and RSS
Exhibit tracking slip
Verdict
Saving a URL, extracting a clean reading view that strips ads and navigation, and full-text search across saved articles is a genuine weekend build using an existing open readability-extraction library — that's most of Instapaper's actual value. What doesn't survive: extraction reliability across the messiest, most unusual sites, and years of format-specific fixes for paywalled sites, JS-rendered content, and unusual article layouts.
Exhibit A — The prompt
Received on31.07.2026Build a read-it-later tool centered on clean article extraction. Use an open readability-extraction library, such as Mozilla's Readability.js, the same library Firefox's own reader mode uses, server-side to strip ads, navigation, and clutter from a saved URL, leaving just the article text and images. Use Next.js/TypeScript with Postgres and its full-text search for searching across all saved articles' extracted text. When a user saves a URL, fetch and extract it immediately, and store both the extracted text and a fallback link to the original in case extraction fails or looks wrong. Build a clean reading view with adjustable font size and light/dark mode, and a list view with read, archived, and unread states and tags. Add a basic bookmarklet or browser extension for one-click saving from any page. Do not attempt text-to-speech, mobile app offline sync, or extraction reliability tuning beyond what the open library already provides — sites that defeat the readability library, heavy JS rendering or paywalls, will sometimes need the fallback original link instead of a clean view, and that's an honest, expected gap. No external API key needed; requires hosting and a database.Opening prefills the prompt — press enter to run it.
Exhibit B — What you lose
- B.1 extraction reliability across unusual or JS-heavy sites
- B.2 years of site-specific parsing fixes
- B.3 browser and mobile app polish with offline sync
- B.4 text-to-speech and speed-reading modes
Prior art
Exhibit C — Why people still pay: capture polish, sync, and content partnerships
Readability extraction libraries handle most sites well; handling the long tail of sites that break the common pattern, reliably, for years, is the maintenance work underneath.
Questions
Will every article extract cleanly?
Most will, since the underlying readability library handles common site patterns well — but some sites, especially JS-heavy or paywalled ones, will fail to extract cleanly, and you'll fall back to the original link.
Can I search across everything I've saved?
Yes — full-text search across extracted article text is a core feature here, using Postgres's built-in search.
Does it work offline on my phone?
Not in this build — there's no native mobile app with offline sync, just a web app. That's one of the real things you give up.
What does it cost to run?
Hosting and a database — typically a few dollars a month, since there's no external extraction API this build depends on.
Related tools
Receipt