audit_page
ChatGPTCheck how ready one web page is to be read and cited by AI assistants: whether robots.txt lets the AI crawlers (GPTBot, ClaudeBot, PerplexityBot, Google-Extended and others) in, llms.txt, the sitemap, structured data, headings, trust signals, meta tags and a Wikidata entry. Returns a 0-100 score, an A-F grade, the score per category, the main issues and the raw signals. Use this tool when the user asks whether a website is ready for AI search, why AI might not cite a page, or what to fix on it -- phrases like 'audit our homepage for AI', 'can ChatGPT crawl our site', 'GEO check of example.com'. It checks the page itself. It does not measure what AI assistants answer about the brand; that needs a Refora workspace. Example question: 'Is kassaro.se ready for AI search?'
get_ai_answers
ChatGPTThe answers themselves: what ChatGPT, Perplexity, Gemini, Claude, Mistral and Google AI Overviews said to this workspace's prompts, each with the brands it mentions and the URLs it cites. Use this tool when the user wants to read or quote what an AI said -- phrases like 'what does ChatGPT say about us', 'show me the answers to the prompt about X', 'how do AI models describe our competitor', 'show answers where we are not mentioned' -- or to explain a number another Refora tool returned. For numbers (visibility, share of voice, trends, sentiment, sources), use those tools: they count every answer, this one returns a page. Filter by prompt (its text or a distinctive part of it), model, brand (brand_mentioned=true for answers that mention it, false for answers that don't; without brand_name it is the workspace's own brand) and window. Answers come newest first, 10 per page; when next_offset is set, call again with offset=next_offset. An answer longer than 2,000 characters is cut (truncated: true); to read it whole, call again with its prompt_text, model, its date as start_date and end_date, and full_text=true. Example: 'What did Perplexity say when asked for the best CRM?'
get_citation_sources
ChatGPTThe domains and URLs that AI models cite in their answers for this workspace's prompts, each marked as the brand's own site, a competitor's, or other. Each URL also says what kind of page it is and how often each model cited it. Use this tool when the user asks which websites or sources AI references -- phrases like 'what sources does AI cite', 'which websites does ChatGPT reference', 'top cited domains', 'which comparison pages does AI cite', or 'where does AI get its information about us'. Prefer it to paging through get_ai_answers for any question about cited pages. With brand_name, only answers that mention that brand count. Without it, every answer in the workspace counts. model, page_type, owner and last_runs narrow what counts, in one call: e.g. model=["openai", "google-aio"], page_type=["comparison", "listicle"], owner="competitor" and last_runs=3 for the competitor comparison and best-of pages ChatGPT and Google AI Overviews cited in the last 3 runs. Every URL carries its page_type and citations_by_model, and per_model gives each model's runs and citations, so a count reads as 'N citations in M runs'. A URL is a page: tracking tags (utm_*) and #fragments are dropped, so a page two models cite is one row. For one topic, use get_top_cited_urls_for_topic. Across every workspace in the organization, use get_top_cited_domains. Example: 'What sources does AI cite most when talking about Kassaro?'
get_competitor_comparison
ChatGPTRank the workspace's own brand against the competitors the user tracks, by visibility: the share of AI answers that mention each brand. Use this tool when the user asks about competitive position -- phrases like 'how do we compare', 'who's winning', 'where do we rank', or 'vs competitors'. Do NOT use for the own score alone (use get_visibility_score), trends (use get_visibility_trend), or every brand the AI mentions, tracked or not (use get_share_of_voice). Returns: tracked brands ranked by visibility, the own brand among them. Brands with nothing to measure get no rank. To see WHY a competitor leads, follow up with get_competitor_visibility_gap. Example: 'How does Kassaro compare to its competitors?'
get_competitor_visibility_gap
ChatGPTThe prompts where one tracked competitor is mentioned in AI answers more often than the workspace's own brand. Use this tool as a follow-up to get_competitor_comparison when the user wants to understand WHY a competitor is winning -- phrases like 'why is X beating us', 'where are we losing to X', 'what sources does AI cite for X', or 'how do we beat X'. Returns: prompts where the competitor leads by at least min_gap_percent percentage points, each with the gap, the AI models that favor the competitor, and the URLs AI cites in those answers (the brand's own pages left out). Use this data for content briefs. Example: 'Why is Sitoo beating Kassaro, and what sources does AI cite for them?'
get_market_ai_presence
ChatGPTA market snapshot of the workspace's website from an external index of AI answers: ChatGPT's and Gemini's answers to the brand's buyer questions as users see them in its country (share of answers naming the brand and each competitor, who else they name, which sources they cite), and how often Google's AI Overviews cite the site next to its competitors, month by month, with the questions where a competitor is cited and the site is not. Use this tool when the user asks how visible they are in AI before their own questions are tracked, or wants the wider market picture next to their own tracking -- phrases like 'does ChatGPT recommend us', 'who does AI recommend instead of us', 'how do we compare in Google's AI answers'. The snapshot is built once, when the account is created; status 'building' means it is not ready yet. Its numbers are not the workspace's own tracking: read each section's about. Example question: 'Does ChatGPT recommend Kassaro?'
get_prompt_performance
ChatGPTDetailed performance for one prompt (one question Refora asks the AI models on a schedule): visibility over time, per-model visibility and answers, and the URLs cited in its answers. Use this tool when the user asks about a specific prompt -- phrases like 'how are we doing on the prompt about X', 'details on this prompt', or as drill-down after seeing prompts from get_competitor_visibility_gap. Name the prompt by its text: the full text as another Refora tool returned it, or a distinctive part of it. Prompts with the same text are one question and are measured together. If several prompts match, the response lists them so the user can pick. Example: 'How are we doing on the prompt about enterprise CRM?'
get_sentiment_breakdown
ChatGPTHow AI models talk about a brand: sentiment scores overall and per model, with example excerpts showing tone and framing. Use when the user asks how AI talks about their brand -- phrases like 'how does AI describe us', 'what's the sentiment', 'is ChatGPT positive or negative about us', or 'how do AI engines talk about [brand]'. Returns: overall sentiment score (0-100, 50 is neutral), per-model breakdown, and up to 3 examples per label (positive/neutral/negative), each a short excerpt around the mention. Example: 'How does ChatGPT describe Kassaro -- positively or negatively?'
get_share_of_voice
ChatGPTThe workspace's own brand's share of ALL brand mentions in AI answers, and which brands hold the rest -- every brand the models named, each marked as a tracked competitor or not. Use this tool when the user asks who the AI talks about -- phrases like 'share of voice', 'who gets mentioned most', 'which brands does AI recommend in our space', or 'who else shows up'. Do NOT use to rank against the competitors the user tracks (use get_competitor_comparison, which measures the share of answers that mention each tracked brand). Returns: own share in percent, own rank, and the top brands with their shares. Example: 'Which brands does AI mention most when people ask about our category?'
get_top_cited_domains
ChatGPTThe domains AI models cite most across ALL workspaces in the organization. Aggregate view for multi-workspace orgs. Use when the user wants org-level source analysis -- phrases like 'across all workspaces', 'org-wide sources', 'top domains overall', or 'which sites dominate'. For one workspace, use get_citation_sources. Example: 'Which domains are cited most across all our brands?'
get_top_cited_urls_for_topic
ChatGPTThe URLs that AI models cite most for the workspace's prompts in one topic. Use this for content briefs -- it shows which sources AI treats as authoritative for that topic. Use when the user asks about sources for a topic -- phrases like 'what does AI cite for X topic', 'sources on X', 'content research for X', or 'authority sources for X'. Topics are the workspace's own; most are one of PROBLEM_AWARENESS, SOLUTION_COMPARISON, COMPETITOR_COMPARISON, EVALUATION, CATEGORY_DEEP_DIVE, HOW_TO. If the topic is not found, the response lists the topics that exist. model, page_type, owner and last_runs narrow what counts, in one call: e.g. model=["openai", "google-aio"], page_type=["comparison", "listicle"], owner="competitor" and last_runs=3 for the competitor comparison and best-of pages ChatGPT and Google AI Overviews cited in the last 3 runs. Every URL carries its page_type and citations_by_model, and per_model gives each model's runs and citations, so a count reads as 'N citations in M runs'. A URL is a page: tracking tags (utm_*) and #fragments are dropped, so a page two models cite is one row. Example: 'What sources does AI cite for our EVALUATION prompts?'
get_visibility_score
ChatGPTHow visible the workspace's own brand is in AI answers: the share of AI answers that mention it, overall and per AI model (ChatGPT, Perplexity, Gemini, Claude, Mistral, Google AI Overviews), and the change from the window before. Use this tool when the user asks about current performance -- phrases like 'how are we doing', 'what's our score', 'how visible are we in AI', or 'current visibility'. Do NOT use for trends over time (use get_visibility_trend) or competitors (use get_competitor_comparison). Returns: visibility in percent (null means no answers to measure, not zero), the change in percentage points, one row per model with its visibility and number of answers, and a summary. Example: 'How is Kassaro doing in AI search right now?'
get_visibility_trend
ChatGPTVisibility over time for the workspace's own brand. Use this tool when the user asks about changes over time -- phrases like 'how has it changed', 'trending', 'are we improving', 'show the trend', or 'over the last month'. Do NOT use for the current snapshot (use get_visibility_score) or competitors (use get_competitor_comparison). Returns one data point per period and a trend summary. A period with no AI answers has a null visibility, not 0. Max 90 days. Example: 'How has our AI visibility changed over the last 30 days?'
list_workspaces
ChatGPTList the workspaces in the connected Refora organization. Call this first: every other workspace tool needs a workspace_id from here. Use this tool when the user asks about their workspaces, projects, or brands they're monitoring -- phrases like 'what am I tracking', 'list my projects', 'show me my brands', or 'which workspaces do I have'. Returns: workspace ID, name, primary brand domain, active prompt count, creation date, website, data_source (own_tracking, market_data or none) and a one-line summary for each workspace. Example question: 'What workspaces do I have set up in Refora?'