fb-scraper

Internal Facebook video scraper service — multi-provider, no API key required.

Live: https://fb.xcasper.space

Share: Facebook · Twitter/X · WhatsApp

Part of: CASPER TECH API Hub — 260+ free API endpoints

---

What Is This?

fb-scraper is a lightweight internal Node.js HTTP service that powers the Facebook video downloader endpoints on the CASPER TECH API Hub (apis.xcasper.space). It runs on the VPS at 127.0.0.1:5757 and is not publicly exposed — all traffic comes from the Next.js API layer.

It provides three scraping strategies, each targeting a different upstream provider:

1. fdown.net (Puppeteer) — navigates fdown.net via a headless Chromium browser, submits the Facebook URL, and extracts fbcdn.net direct download links. HD/SD detected via URL pattern (/m366/ = HD, /m412/ = SD).
2. savefrom.net (Puppeteer) — drives en1.savefrom.net with a stealth browser session, extracts quality-labelled download links and thumbnail.
3. snapsave.app (vm-decode, no browser) — posts to snapsave.app/action.php via plain HTTP, receives obfuscated JavaScript, executes it inside a Node.js vm sandbox with a mocked document.getElementById setter, and parses the decoded HTML for d.rapidcdn.app JWT-proxied download links. No Puppeteer required — ~250ms response time.

Built and maintained by TRABY CASPER under the CASPER TECH umbrella.

---

Owner & Author

Name	TRABY CASPER
Organisation	CASPER TECH
Country	Kenya
Website	xcasper.space
GitHub	@Casper-Tech-ke
Role	Founder & Lead Developer

CASPER TECH is a Kenyan tech initiative building free, accessible developer tools and APIs for African and global developers. fb-scraper is a core internal service of the CASPER TECH API Hub — apis.xcasper.space.

---

Endpoints

All endpoints accept the Facebook URL via ?url= query parameter.

Method	Path	Provider	Strategy
GET	/health	—	Health check and uptime
GET	/scrape?url=FB_URL	fdown.net	Puppeteer (Chromium headless)
GET	/scrape-sf?url=FB_URL	savefrom.net	Puppeteer (Chromium headless)
GET	/scrape-snap?url=FB_URL	snapsave.app	Pure HTTP + vm sandbox

---

Self-Hosting

Prerequisites

• Node.js 18 or higher
• Chromium installed (for Puppeteer endpoints)
• PM2 (recommended for production)

Install & Run

# Clone the repo
git clone https://github.com/Casper-Tech-ke/fb-scraper.git
cd fb-scraper

# Install puppeteer-core (and node-fetch if not on Node 18+)
npm install puppeteer-core node-fetch

# Set your Chromium path in server.js (default: /usr/bin/chromium)
# CHROMIUM = '/usr/bin/chromium'

# Run directly
node server.js

# Or run with PM2
pm2 start server.js --name fb-scraper
pm2 save

The service listens on 127.0.0.1:5757 by default. To change the port, update PORT at the top of server.js.

Chromium Setup (Ubuntu/Debian)

apt-get install -y chromium-browser
# or
apt-get install -y chromium

Deploy on Render

1. Go to render.com → New Web Service
2. Connect GitHub → select Casper-Tech-ke/fb-scraper
3. Configure:

Setting	Value
Environment	Node
Build Command	npm install puppeteer-core node-fetch
Start Command	node server.js
Port	5757

Puppeteer note: Render free tier does not ship Chromium by default. Add the Puppeteer buildpack in Settings → Buildpacks. Without it, only /scrape-snap (no browser) will work.

# Environment variable (Render dashboard)
CHROMIUM_PATH=/usr/bin/google-chrome-stable

Deploy on Koyeb

1. Go to koyeb.com → Create App
2. Select GitHub → Casper-Tech-ke/fb-scraper
3. Configure:

Setting	Value
Run Command	node server.js
Port	5757
Plan	Nano (free) — snap endpoint only / Micro for full Puppeteer

Docker deploy (recommended for Puppeteer): Use a Dockerfile that installs Chromium. Example:

FROM node:20-slim
RUN apt-get update && apt-get install -y chromium --no-install-recommends
WORKDIR /app
COPY . .
RUN npm install puppeteer-core node-fetch
ENV CHROMIUM_PATH=/usr/bin/chromium
EXPOSE 5757
CMD ["node", "server.js"]

Deploy on Replit

1. Go to replit.com → Create Repl
2. Select Import from GitHub → paste https://github.com/Casper-Tech-ke/fb-scraper
3. In the Shell tab:

npm install puppeteer-core node-fetch

4. Create or update .replit:

run = "node server.js"

⚠️ Important: Chromium/Puppeteer is NOT available on Replit free tier. Only /scrape-snap (snapsave.app vm-decode — no browser) will work. To use /scrape and /scrape-sf, upgrade to Replit Core.

---

Example Requests

# Health check
curl "http://127.0.0.1:5757/health"

# fdown.net scraper (HD + SD fbcdn.net links)
curl "http://127.0.0.1:5757/scrape?url=https://www.facebook.com/reel/1234567890"

# savefrom.net scraper
curl "http://127.0.0.1:5757/scrape-sf?url=https://www.facebook.com/reel/1234567890"

# snapsave.app — fast, no browser needed
curl "http://127.0.0.1:5757/scrape-snap?url=https://www.facebook.com/reel/1234567890"

---

Response Format

/scrape and /scrape-sf

{
  "success": true,
  "title": "Facebook Video",
  "thumbnail": "https://scontent.fbcdn.net/...",
  "links": [
    { "url": "https://video-xxx.fbcdn.net/...m366...", "label": "HD Download" },
    { "url": "https://video-xxx.fbcdn.net/...m412...", "label": "SD Download" }
  ]
}

/scrape-snap

{
  "success": true,
  "title": "Facebook Video",
  "thumbnail": "https://d.rapidcdn.app/thumb?token=...",
  "links": [
    {
      "label": "720p (HD)",
      "url": "https://d.rapidcdn.app/v2?token=...",
      "isHD": true,
      "isSD": false
    }
  ]
}

Error Response

{
  "error": "Description of what went wrong"
}

---

Quality Detection

For /scrape (fdown.net), HD/SD quality is detected from the fbcdn.net URL pattern:

URL Pattern	Quality
/m366/	HD
/m412/	SD
/m420/	HD
hd in URL	HD
sd in URL	SD

---

How the vm-decode Works (/scrape-snap)

snapsave.app returns obfuscated JavaScript instead of plain HTML. The decoding flow:

1. POST https://snapsave.app/action.php?lang=en with multipart/form-data body url=<FB_URL>
2. Response is ~23KB of obfuscated JS (eval-based)
3. A Node.js vm.Script is created with a mocked browser environment:
   • window.location.hostname = 'snapsave.app'
   • document.getElementById() returns an object with an innerHTML setter that captures the decoded HTML
   • document.querySelector() and remove() stubs to prevent errors
4. The captured HTML is parsed with regex for href="https://d.rapidcdn.app/v2?token=..." download links and thumbnail URLs
5. Links are JWT-proxied fbcdn.net URLs with ~60 minute expiry

---

Supported Facebook URL Formats

https://www.facebook.com/videos/1234567890
https://www.facebook.com/reel/1234567890
https://web.facebook.com/reel/1234567890
https://m.facebook.com/watch/?v=1234567890
https://fb.watch/xxxxxxxx
https://www.facebook.com/share/v/xxxxxxxx
https://www.facebook.com/share/r/xxxxxxxx

---

Dependencies

Dependency	Purpose
puppeteer-core	Headless Chromium automation for fdown.net and savefrom.net
node-fetch	HTTP requests for snapsave.app
Built-in vm	Sandboxed JS execution for snapsave obfuscation decode
Built-in http	HTTP server

Chromium path: /usr/bin/chromium Puppeteer path: /tmp/iss-capture/node_modules/puppeteer-core

---

Project Structure

fb-scraper/
├── server.js              - Main HTTP server (all scraping logic + static file serving)
├── server.js.bak          - Backup of base server before snap function was added
├── public/
│   ├── index.html         - Interactive documentation frontend
│   ├── terms.html         - Terms of Service
│   ├── privacy.html       - Privacy Policy
│   ├── disclaimer.html    - Disclaimer page
│   ├── favicon.svg        - SVG favicon
│   └── og.png             - Social preview image (1200×630)
├── CONTRIBUTING.md
├── DISCLAIMER.md
├── LICENSE
└── README.md

---

API Hub Integration

This service powers the following public endpoints on apis.xcasper.space:

Public Endpoint	Uses	Strategy
POST /api/downloader/fb	/scrape	fdown.net Puppeteer, fallback: xaviabot
POST /api/downloader/fb2	/scrape-sf + /scrape	savefrom.net → fdown.net → xaviabot
POST /api/downloader/fb3	/scrape	fget.io direct API → fdown.net → xaviabot
POST /api/downloader/fb4	/scrape-snap	snapsave.app vm-decode

---

Contributing

Contributions are welcome! Please read CONTRIBUTING.md before submitting a pull request.

Legal

• Disclaimer — not affiliated with Facebook/Meta
• Terms of Service — usage rules and restrictions
• Privacy Policy — no personal data stored
• DISCLAIMER.md — GitHub markdown version

License

MIT © 2026 TRABY CASPER · CASPER TECH

---

Built with passion in Kenya by TRABY CASPER · CASPER TECH