Click any run from Visibility → Runs or the Inbox to land on its Run Detail page. The page is one long scroll with several sections.
Status header
Top of the page shows the run status: QUEUED → GENERATING_PROMPTS → AWAITING_REVIEW → RUNNING_PROMPTS → ANALYZING_GAPS → GENERATING_DRAFTS → COMPLETED. For runs in progress, you’ll see a progress count (e.g. 14 / 30 probes) so you can tell whether it’s 60% done or just sitting at “running.”
Three Key Findings
Right below the header, the synthesised findings — three plain-English statements about what changed and what to do next. Each finding has a recommended actions list you can approve one-click into the Inbox.
Prompt grid
One row per prompt, columns per engine. Each cell shows whether you were cited, where (position), and the sentiment classification. Sortable by self-citation rate. Click a row to drill into the per-prompt detail page with the full LLM response.
Citations table
Every domain cited across the run, ranked by frequency. Your domain is highlighted; tracked competitors are marked. Click a domain to see which prompts cited it.
Filter decisions (collapsible)
The naturalness filter (Step B) sometimes rejects generated prompts that don’t read like real buyer questions. The collapsible at the bottom shows what got rejected and why — useful for tuning if you think the agent is being too strict.