Results
Use the restored website ZIP and read its restoration report.
Output files
Completed restorations produce a ZIP and a JSON report. By default, they are
stored in data/output in the directory where you run the command:
data/output/jobs/<job-id>/
├── snapshot.zip
└── result.jsonThe dashboard uses the same layout and provides a Download ZIP link. See Storage to choose other directories.
Website archive
snapshot.zip contains the restored pages and assets, with HTML and CSS
references rewritten to their local paths. Files are stored relative to the
website root. Extensionless pages use directory indexes such as
about/index.html. Query-string variants get distinct filenames.
The replay URL you restore is the entry page and is saved as the ZIP's root
index.html, including when it was a subpage such as /html/index/index.php.
Its links keep the original routes, and links to the site root open it.
Extract the ZIP and serve the site locally, or upload the extracted files to a static web host.
Job report
result.json records the requested capture, resource counts, selected fallback
captures, plugin counters, and archive metadata. It stays alongside the ZIP.
The engine writes this report after processing resources, including bounded
runs and runs whose entry page could not be restored.
A completed-job report includes fields such as:
{
"event": "final_result",
"status": "completed",
"job_id": "example-september",
"domain": "example.com",
"capture_timestamp": "20260925084419",
"counts": {
"pending": 0,
"processing": 0,
"completed": 12,
"unavailable": 0,
"failed": 0
},
"failure_count": 0,
"failures": [],
"artifact": "snapshot.zip"
}Job status
status | Meaning |
|---|---|
completed | All discovered resources restored; ZIP available. |
completed_with_source_gaps | Some resources other than the entry page were unavailable or failed; ZIP available. |
bounded | --max-files was reached with resources still pending; report available. Continue the job to create a ZIP. |
failed | The entry page was not restored; error says why. |
ZIPs are published for completed and completed_with_source_gaps jobs.
Resource counts
| State | Meaning |
|---|---|
pending | Waiting to be processed. |
processing | Currently being processed. |
completed | Downloaded and available for the restored output. |
unavailable | Recorded as unavailable from the archive. |
failed | Processing ended with an error. |
Report fields
| Field | Meaning |
|---|---|
job_id, domain, capture_timestamp | Job and requested capture. |
error | Why the entry page was not restored; null unless status is failed. |
counts | Resources grouped by the states above. |
failure_count, failures | Total unavailable/failed resources; each entry has original_url, state, error, and attempts. |
failure_samples | First 50 failures; also shown in the CLI final event. |
linked_page_capture_fallback_count, linked_page_capture_fallbacks | Pages restored from older linked-page captures; each entry has url, discovered_from, capture_timestamp, and drift_days. |
files_materialized, remote_neutralized | Files written and remote references disabled during rewriting. |
plugins | Enabled plugins and their counters. |
artifact, artifact_size, artifact_sha256, file_count, extracted_size | ZIP name, compressed bytes, SHA-256, file count, and uncompressed bytes; present when a ZIP is created. |
Capture details
capture_timestamp is the requested capture date. In replay mode,
linked_page_capture_fallbacks lists linked pages recovered from older captures
beyond the normal page window. Each entry records the page URL, the page that
discovered it, the selected timestamp, and its distance from the requested date.
See Capture selection to configure these windows.
Plugin results
Each entry in plugins contains the invocation's id, registered plugin name,
and counters. For example, path_prefix reports renamed paths and rewritten
references. See Plugins for the available transformations.