Quickstart

InstallCopy Anchor Link

Install with pipx or uv. Both put the wayback-restorator command in an isolated environment. It requires Python 3.11 or newer:

pipx install wayback-restorator
# or
uv tool install wayback-restorator

Upgrade later with pipx upgrade wayback-restorator or uv tool upgrade wayback-restorator.

Use Docker insteadCopy Anchor Link

Without Python, run the published image with Docker. Replace wayback-restorator in the commands below with:

mkdir -p data
docker run --rm -it --user "$(id -u):$(id -g)" -v "$PWD/data:/app/data" \
  registry.gitlab.com/restorator-labs/wayback-restorator

The container stores its files in data/ in your current directory. For the dashboard, also publish its port; see Dashboard.

To work on the source code, see Development.

Choose a captureCopy Anchor Link

Open the website in the Wayback Machine, select a capture, and copy its full URL from your browser. For example:

https://web.archive.org/web/20260925084419/https://example.com/

• 20260925084419 is the capture timestamp: September 25, 2026, at 08:44:19 UTC.
• https://example.com/ is the original page address.

Use a URL containing the full 14-digit timestamp.

Restore with the dashboardCopy Anchor Link

wayback-restorator web

1. Open localhost:8080.
2. Paste the capture URL into Wayback replay URL.
3. Click Add to queue.
4. Follow the job's progress and click Download ZIP when it finishes.

See Dashboard for the queue and settings.

Restore from your terminalCopy Anchor Link

wayback-restorator restore 'https://web.archive.org/web/20260925084419/https://example.com/'

The terminal shows restoration progress and prints the paths to the finished archive and report:

data/output/jobs/<job-id>/snapshot.zip
data/output/jobs/<job-id>/result.json

See CLI for options, named jobs, and resuming.

Open the restored websiteCopy Anchor Link

Extract the ZIP into a directory named restored-site. Serve that directory using Python:

python3 -m http.server 8000 --bind 127.0.0.1 --directory restored-site

Open localhost:8000. The selected entry page is saved as the root index.html, including when the original URL was a subpage.

See Results for the archive layout and report fields.