Configuration
Capture selection, crawl scope, requests, and storage settings.
Configure CLI restorations with options:
wayback-restorator restore 'https://web.archive.org/web/20260925084419/https://example.com/' \
--concurrency 4 --max-pages 100In the dashboard, use the corresponding form fields. The dashboard saves those settings with each queued job.
Environment variables
Every CLI option has an environment variable named after it: --max-pages
becomes MAX_PAGES, and the capture URL is WAYBACK_URL. Options passed on the
command line take precedence. Without a command, wayback-restorator restores
the capture configured in the environment:
WAYBACK_URL='https://web.archive.org/web/20260925084419/https://example.com/' \
CONCURRENCY=4 wayback-restoratorCapture selection
| CLI option | Dashboard field | Default | Values / meaning |
|---|---|---|---|
--capture-mode | Capture mode | replay | replay or strict. |
--max-page-capture-drift-days | Page drift (days) | 120 | Non-negative number of days between the requested capture and a page capture. |
--max-linked-page-capture-drift-days | Linked page drift (days) | 365 | Older linked-page fallback window in replay mode; at least the page drift value. |
--max-asset-capture-drift-days | Asset drift (days) | 3650 | Non-negative asset capture window used in strict mode. |
Replay mode
Replay is the default. Pages use the page drift window. For a discovered linked page, the engine can also use an older capture within the linked-page window. Assets are selected from the available capture candidates across dates.
The report records linked pages recovered through the older-capture fallback, including their actual timestamps.
Strict mode
Strict mode applies the page drift window to pages and the asset drift window to assets. Select it when you want captures within specific date windows:
wayback-restorator restore 'https://web.archive.org/web/20260925084419/https://example.com/' \
--capture-mode strict --max-page-capture-drift-days 30 --max-asset-capture-drift-days 90Crawl scope
| CLI option | Dashboard field | Default | Values / meaning |
|---|---|---|---|
--max-pages | Max pages | Unlimited | Positive integer; includes the starting page. 0 means unlimited. |
--max-depth | Max depth | Unlimited | Positive integer; the starting page has depth 0. 0 means unlimited. |
--max-files | Max files | Unlimited | Positive integer; counts resources reaching completed, unavailable, or failed states. |
--cross-origin-assets | Include cross-origin assets | Off | Include assets from other origins through Wayback. Turn off with --no-cross-origin-assets. |
The dashboard uses empty fields for unlimited values. Page and depth settings control page discovery; assets referenced by selected pages are still discovered. File bounds count all resources, including pages and assets, across the saved job.
A job with a restored entry page stopped by a file bound with resources still
pending is bounded. It writes a report; continuing it to completion creates
the ZIP. A page or depth bound defines the selected crawl scope, which can
produce a completed ZIP.
Requests
| CLI option | Dashboard field | Default | Values / meaning |
|---|---|---|---|
--concurrency | Concurrent requests | 6 | Integer from 1 to 32. |
--request-max-attempts | Request attempts | 3 | Positive integer; total attempts for a transient request error. |
--retry-base-delay-seconds | Retry base (seconds) | 1 | Non-negative number; initial retry delay. |
--retry-max-delay-seconds | Retry max (seconds) | 30 | Non-negative number, at least the base delay. |
Retries use an exponential delay capped by the maximum retry delay for
transient network errors and retryable Wayback responses. Capture windows and
retry delays must be below 100000.
Proxy
Set the PROXY_URL environment variable to route requests from the CLI and the
dashboard through an HTTP or HTTPS proxy:
PROXY_URL=http://proxy.example:3128 \
wayback-restorator restore 'https://web.archive.org/web/20260925084419/https://example.com/'Credentials can be included in the URL, for example
http://user:password@proxy.example:3128. A value without a scheme uses http://.
An empty value uses a direct connection. The proxy has no command-line option,
so credentials stay out of your shell history.
Storage
| CLI option | Environment variable | Default | Contents |
|---|---|---|---|
--work-dir | WORK_DIR | data/work | Job state and downloaded resources. |
--output-dir | OUTPUT_DIR | data/output | Reports and ZIPs. |
--job-id | JOB_ID | Derived from the capture URL | Job directory name: 1–180 letters, digits, dots, underscores, tildes, or hyphens. |
Relative paths start from the directory where you run the command. Both
restore and web accept the directory options; the dashboard generates its
own job IDs. Keep the working files to continue saved jobs.
The Docker image uses /app/data/work and /app/data/output. Mount a host
directory or volume at /app/data to keep them.
Dashboard server
| CLI option | Environment variable | Default | Purpose |
|---|---|---|---|
--host | WEB_HOST | 127.0.0.1 | Address the dashboard listens on. |
--port | WEB_PORT | 8080 | Port the dashboard listens on. |
Keep the default host unless the dashboard runs in a container. It has no authentication.
Output plugins
--plugins-json takes a JSON array of plugin configurations. Its default is
[]. In the dashboard, paste the array into Output plugins JSON.
wayback-restorator restore 'https://web.archive.org/web/20260925084419/https://example.com/' \
--plugins-json '[{"id":"clean","plugin":"strip_external_urls","options":{}}]'See Plugins for available plugins, options, and custom transformations.