RSS → EPUB

A single-page, client-side tool that turns any RSS or Atom feed into EPUB files for reading on an e-ink device. Paste a feed URL, pick your options, and download either one combined EPUB of all articles or one EPUB per article (zipped). Everything runs in the browser — the only moving part is a tiny Cloudflare Worker that proxies the network requests.

Features

• Full-text articles, not just feed excerpts — fetches each article page and extracts the readable content with Mozilla's Readability.
• Combined or individual output: one book with every article as a chapter, or a zip of separate EPUBs.
• Images included and cleaned up — embeds article images directly in the EPUB, filters out ads/trackers and tracking pixels, handles lazy-loaded images, and converts WebP/AVIF to JPEG for broad e-reader compatibility. Can be turned off for fast, text-only books.
• Cleaner articles — strips common end-of-article clutter: "Add as a preferred source on Google" / Google News promos, FTC affiliate disclosures, and author social-media follow lines (in-article product links are left intact).
• Cover + table of contents in combined mode: a cover image (feed logo, or the first article's image, with a text title page as fallback), a title/metadata page, and a clickable Contents page, plus the reader's built-in navigation.
• Byline + source link on each article — author and date when the feed provides them, plus a "Read original" link back to the source page.
• Fetch controls: cap the number of articles, limit to the last N days, and skip articles you've already downloaded (remembered per feed in your browser).
• Saved feeds: keep a dropdown of your regular feeds, remembered in the browser, so you never retype a URL.

How it works

feed URL ─▶ fetch (via Worker) ─▶ parse RSS/Atom items
                                       │
                 for each item ─▶ fetch full page ─▶ Readability extract
                                       │                    │
                                       ▼                    ▼
                                 embed + clean images   valid XHTML
                                       │
                 ┌─────────────────────┴─────────────────────┐
            one combined EPUB                         one EPUB per article
            (cover + TOC + chapters)                  (zipped)
                 └─────────────────────┬─────────────────────┘
                                       ▼
                                   download

Libraries (JSZip and @mozilla/readability) load from a CDN at runtime, so there's no build step — index.html is the entire app.

Setup

1. Host the page (GitHub Pages)

1. Add index.html to a repository.
2. Settings → Pages → Source: Deploy from a branch → main / root.
3. Open the Pages URL GitHub gives you.

2. Deploy the proxy (Cloudflare Worker)

A proxy is required because browsers block cross-origin requests, so the page can't fetch feeds or article pages directly. A personal Cloudflare Worker handles this (free tier is 100,000 requests/day — far more than you'll need).

1. dash.cloudflare.com → Workers & Pages → Create → Create Worker.

2. Name it (e.g. rss-proxy), Deploy, then Edit code and replace the default with:

   export default {
     async fetch(request) {
       const CORS = {
         "Access-Control-Allow-Origin": "*", // or your Pages origin, e.g. https://you.github.io
         "Access-Control-Allow-Methods": "GET, OPTIONS",
         "Access-Control-Allow-Headers": "*",
       };
       if (request.method === "OPTIONS") return new Response(null, { headers: CORS });
   
       const target = new URL(request.url).searchParams.get("url");
       if (!target) return new Response("Missing ?url=", { status: 400, headers: CORS });
   
       let dest;
       try { dest = new URL(target); }
       catch { return new Response("Bad URL", { status: 400, headers: CORS }); }
       if (dest.protocol !== "http:" && dest.protocol !== "https:")
         return new Response("Only http/https allowed", { status: 400, headers: CORS });
   
       try {
         const upstream = await fetch(dest.toString(), {
           headers: {
             "User-Agent": "Mozilla/5.0 (compatible; RSS2EPUB/1.0; personal reader)",
             "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8",
           },
           redirect: "follow",
         });
         const body = await upstream.arrayBuffer();
         const headers = new Headers(CORS);
         headers.set("Content-Type", upstream.headers.get("Content-Type") || "text/plain; charset=utf-8");
         return new Response(body, { status: upstream.status, headers });
       } catch (e) {
         return new Response("Upstream fetch failed: " + e.message, { status: 502, headers: CORS });
       }
     }
   };

3. Deploy and copy the Worker URL (e.g. https://rss-proxy.your-subdomain.workers.dev).

3. Point the page at your Worker

In index.html, set the proxy line near the top of the <script> to your Worker URL, keeping /?url= on the end:

const PROXY = "https://rss-proxy.your-subdomain.workers.dev/?url=";

Commit, and you're done.

Usage

1. Paste a feed URL (e.g. https://example.com/feed.xml).
2. Set Max articles and, optionally, a last N days window.
3. Choose combined or one per article, and toggle Include images.
4. Click Generate. The EPUB (or zip) downloads to your machine; copy it to your e-reader over USB.

Skip already-downloaded remembers article IDs per feed in localStorage, so repeat runs only grab new pieces. Use Reset history to clear that memory for the feed currently in the box.

Saved feeds — click Save to add the current URL to the Saved feeds… dropdown (labeled by site name); pick one anytime to fill the box, or Remove to delete the selected feed. The list is stored in your browser.

Transferring to your device

EPUB is the standard e-ink format, so just copy the downloaded file to your reader like any other book (USB, or whatever method your device uses).

Limitations

• Extraction quality varies by site. Most feeds come out clean; some will have stray navigation text or an imperfect image or two.
• Paywalled articles only yield what a logged-out visitor would see.
• Image-heavy runs are slower, since each image is a separate request through the Worker.
• The ad filter is good, not perfect — a site serving ad images from its own domain with innocuous names could occasionally slip one through.
• Dedupe and saved feeds are per-browser — running on a phone and a laptop keeps separate histories and feed lists.

Hardening (optional)

By default the Worker allows any origin (Access-Control-Allow-Origin: "*"). To lock it to only your own site, set that header to your Pages origin instead, e.g. https://your-username.github.io (no trailing slash).

Security & privacy

Runs entirely in your browser; no server stores anything. Feed and article requests pass through your own Cloudflare Worker, and downloaded EPUBs never leave your machine.

License

Released under the MIT License.