list-crawl-reports

List saved crawl reports

This report is the index for local crawl history. It returns metadata only, ordered newest first, so an agent can choose a snapshot without loading pages and issues or fetching the site again. It says what is stored locally, not what is currently true online.

Run it

For MCP, call seo_list_reports with category crawl if discovery is needed. Call seo_describe_report with { "id": "list-crawl-reports" }, then seo_run_report with:

{
  "id": "list-crawl-reports",
  "params": {
    "site": "sc-domain:example.com",
    "limit": 10
  }
}

Confirm MCP isError is false, then consume structuredContent. The equivalent CLI commands are:

seo reports describe list-crawl-reports --json
seo reports run list-crawl-reports --params '{"site":"sc-domain:example.com","limit":10}' --json

The site filter prevents an unrelated recent crawl from becoming the implicit source. Increase limit only when older baselines are genuinely needed.

Choose a source deliberately

Review each report's id, site, start URL, creation time, status, total pages, issue count, configuration hash, and storage version. Counts are orientation metadata, not enough to judge severity or data completeness. A smaller page count could reflect a scope change, cap, failure, or real site change.

Select an explicit id and load it with get-crawl-report before analysis. For a comparison, inspect both candidates' configuration and caveats instead of assuming adjacent timestamps are comparable. If the expected snapshot is absent, do not silently substitute another site or a fresh crawl; state that the local report was not found. Report lists can also expose old data, so include the creation time whenever presenting a saved finding as context.