Changelog

• get_traffic_attribution_overview: the response now carries rows_note, which names any days that are in the totals but not in the rows. A partial first bucket is counted in totals only.
• get_page_traffic, get_ai_crawl_summary, get_ai_crawl_urls and get_ai_revenue_pages: descriptions now say what is counted and what is left out. For example, page-level totals can sit a few percent below the channel-level overview, and the daily crawl chart covers only the five busiest bots.
• get_cited_domains, get_url_report, get_owned_content_recommendations and get_earned_content_recommendations: a value outside an allowed list, such as an unknown sort, status or priority, now returns an error that lists the accepted values. Before, it was ignored or returned an empty list.
• get_technical_recommendations: returns 2 pages by default (was 5), and leaves out the fixed explanation text unless include_explanations is true.
• get_engine_visibility_trend, get_competitor_visibility and get_prompt_visibility_trend: dates are returned as YYYY-MM-DD, and trend granularity as day, week or month.
• get_prompt_top_brands and get_prompt_metrics: descriptions now say which answers share of voice is measured across.

• get_page_trafficHuman visits per page, split into organic, direct, AI referral and other, with key events and the first and last day seen.
• get_search_performanceSearch Console clicks, impressions, click-through rate and position per page.

• list_urls: now lists the URLs known to the brand's Search Console sync, and adds AI-citation fields (cited, citedPromptCount, citedBy) and a cited filter. It no longer offers url as a sort column.
• get_attribution_health: new coverage block that says what each source can tell you. Google Analytics is the only source of human visits; Cloudflare, Vercel and WordPress report AI bot requests only.
• AI crawler tools: descriptions now say that hits are requests by AI bots, not human visits.

• list_domain_topicsA brand's prompt topics, selected and suggested.
• get_source_run_daysThe days a brand's prompts were run and their cited sources recorded.

• get_probe_run: no longer returns internal cost and credit fields.
• get_url_report and get_cited_domains: each metric (retrievals, retrieved_pct, retrieval_rate, citation_rate) is now defined, so a rate such as 1.82 is not read as a percentage.
• get_competitor_visibility: share of voice is now worked out from each brand's mentions, in line with the Competitors page. Each brand reports its mentions and the basis used.

• get_traffic_attribution_overviewVisits by channel over time, with window totals and the previous window.
• get_traffic_attribution_sourcesConnection state of each data source behind Traffic Attribution.
• get_agent_visit_addonWhether the AI crawler add-on is on, and how much of it is used.
• get_agent_visit_recommendationsCrawl-health fixes found from AI crawler visits.
• get_attribution_healthOne-call health check of the data behind the attribution dashboards.

• get_ai_crawl_summaryAI crawler visits: hits, active bots, failure rate and top pages.
• get_ai_crawl_urlsThe pages AI crawlers fetch, per URL.
• get_ai_revenue_summaryHeadline AI revenue numbers over a window.
• get_ai_revenue_trendAI sessions, conversions or revenue over time.
• get_ai_revenue_pagesLanding pages that receive AI-referred traffic.

• get_ai_visibility_setup_statusWhether setup is finished and whether the trial has expired.
• get_competitor_metricsLatest visibility, share of voice and average rank for competitor brands.
• get_prompt_metricsVisibility, rank and share of voice per tracked question and topic over a date window.
• get_cited_domainsCitation metrics for every website AI answers drew on.
• get_probe_runStatus and progress of one AI-visibility probe run.
• get_probe_resultOne AI engine's full answer to one tracked question.
• get_probe_scheduleProbe and site-audit schedules: cadence, last and next run.

Read-only views that match those tabs in the ThriveStack app.

• get_url_report: adds a date window, sorting, and filters for domain or URL type, brand and source class.
• get_competitor_visibility: adds start_date and end_date.
• get_traffic_attribution_sources: rows can carry an issue (error, quiet, stale or disconnected) that explains why a source needs attention.

• get_domain_promptDetail of one tracked prompt.
• get_prompt_chatsThe AI answers recorded for one prompt.
• get_prompt_chatOne recorded AI answer, with the brands it names and its citations.
• get_prompt_sourcesThe sources AI answers used for one prompt.
• get_prompt_top_brandsThe brands AI answers name for one prompt, ranked.
• get_prompt_visibility_trendOne prompt's visibility over time.

• get_content_audit_summaryContent audit status and page-audit limits.
• list_content_audit_pagesThe pages tracked for the content audit.
• list_content_audit_sitemap_urlsSitemap URLs that can be added to the audit.
• get_brand_kitThe saved brand voice, rules and naming.

• get_domain_prompts: adds status and topic filters and paging.
• get_technical_recommendations: adds a workflow filter and paging.
• get_domain_report: adds a topic filter and paging.
• get_prompt_research and list_prompt_research: add paging.
• get_owned_content_recommendations and get_earned_content_recommendations: descriptions now cover status and assignees.