Command
Use my supplied host, refreshed URL list, refresh dates, baseline export, and days of 7, 30, or 90. Verify visibility with list_projects. Read list_sources({host,kind:"urls",days,limit:100}), get_agent_analytics({host,section:"crawl",days}), and get_overview({host,days}). Match refreshed URLs against returned records and compare compatible measures with my saved export. Report observed changes, unmatched URLs, coverage limits, and whether the supported window overlaps the refresh. Do not request custom date windows, infer retrieval from crawl alone, or claim the refresh caused a change. Return a refresh review with evidence and follow-up checks; missing baseline comparisons remain unknown.What you need
- Provide refreshed URLs, refresh dates, and a comparable saved baseline export.
- Supply a visible host and a 7, 30, or 90 day window.
Keep crawling and citation evidence separate
Match the refreshed pages to returned citation URLs and available crawl records. The review distinguishes a crawler visit from a cited answer, showing which observations support each finding and where the available data cannot establish whether a page was actually retrieved.
Review change without inventing causality
Compare compatible measures from your saved baseline and the current supported window. Document refresh timing, overlapping periods, and missing records so the team can judge the observed change and choose further checks without treating timing alone as proof that the refresh worked.