Getting Started

Run html2rss-web locally with Docker, create one feed from a page URL, then move to production setup when you want stable hosting.

What you need

• Docker
• About two minutes for first success

Step 1: Folder and compose file

mkdir html2rss-web && \
cd html2rss-web

Use the quickstart stack from docker-compose.quickstart.yml. Save it as docker-compose.yml:

services:
  html2rss-web:
    image: html2rss/web:1
    ports:
      - "4000:4000"
    environment:
      RACK_ENV: development
      AUTO_SOURCE_ENABLED: "true"
      HTML2RSS_ACCESS_TOKEN: CHANGE_ME_ADMIN_TOKEN
      BOTASAURUS_SCRAPER_URL: http://botasaurus:4010

  botasaurus:
    image: html2rss/botasaurus-scrape-api:latest

That stack runs:

• html2rss-web on port 4000
• companion botasaurus for JavaScript-rendered pages
• AUTO_SOURCE_ENABLED=true so URL-to-RSS works without YAML
• demo access token CHANGE_ME_ADMIN_TOKEN

Step 2: Start

docker compose up -d

Step 3: Create a feed

1. Open http://localhost:4000 (or http://<LAN_IP>:4000/ on your network).
2. Paste a listing URL — newsroom, changelog, releases, category, or archive page (not a bare homepage).
3. When prompted, enter CHANGE_ME_ADMIN_TOKEN.
4. Open the generated feed URL (.rss or .json).

Next

1. Deployment — production compose, real tokens, LAN HTTP vs HTTPS proxy.
2. Feed Directory — subscribe when a curated feed already covers the site.
3. Advanced Feeds — custom YAML only when auto-source needs tighter selectors.

Ruby CLI (optional)

Without the web UI, one-shot auto-source from the gem:

html2rss scrape https://example.com/blog

html2rss auto is a Thor alias for scrape. Strategy details: Strategy reference.