> ## Documentation Index > Fetch the complete documentation index at: https://docs.hibonsai.com/llms.txt > Use this file to discover all available pages before exploring further. # Agent Visitors > Explore how Instinct, Muse, ChatGPT, and other AI platforms request your storefront's pages. Agent Visitors shows how automated visitors interact with your site. Use it to understand which platforms request your content, how that activity changes, and which parts of your storefront receive the most attention. For ecommerce and marketing teams, this helps answer practical questions: which product and collection pages are agents requesting, and what product information is available on those pages? ![Clean Shot 2026 10 21 At 2 37 56 Pm@2x](https://mintcdn.com/bonsai-fa04a9b1/Vm-GQMes7woeY2aC/images/CleanShot-2026-10-21-at-2.37.56-PM@2x.png?fit=max\&auto=format\&n=Vm-GQMes7woeY2aC\&q=85\&s=96f186376976fb6904be6781858e6771) ## Start with these questions **Requires server-side traffic data.** In order to take full advantage of Agent Visitors, connect the Bonsai edge or a CDN log feed to observe bot activity. If no data has arrived in 30 days, the page shows **Bot traffic is not observed for this store**. Contact your **Customer Success Manager** to set up a feed. > **Which platforms are accessing our site?** > > Use Bot type to compare request volume across ChatGPT, Perplexity, Googlebot, and the other platforms shown. > **Which crawlers are behind the activity?** > > Expand a platform row to inspect the crawler names available. Different bots serve different purposes, from search discovery to fetching a page for a user's request. > **Is agent activity growing or changing?** > > Review Bot traffic over time. Compare request counts alongside shares to see whether a platform's activity changed or its share shifted as other traffic moved. > **Which parts of our site receive the most requests?** > > Review Most crawled paths to find the product and collection pages receiving activity. ## Common bots and platforms you can see **Bot type** groups activity by platform or operator. Expand a row to see individual crawlers and compare their requests, share, and activity over time. | Platform | Bots or activity to look for | What you can explore | | - | - | - | | **OpenAI / ChatGPT** | OAI-SearchBot, ChatGPT-User, GPTBot, OAI-AdsBot | Search crawling, user-requested retrieval, potential training collection, and ad-page review, respectively. | | **Instinct** | Requests identified as Instinct, where available | Requested product and collection pages, request volume, and changes over time. | | **Muse** | Requests identified as Muse, where available | Requested product and collection pages, request volume, and changes over time. | | **Gemini** | Activity grouped under Gemini in your report | Requested pages and trends in activity attributed to Gemini. | | **Claude** | Claude-SearchBot, Claude-User, ClaudeBot | Search crawling, user-requested retrieval, and potential training collection. | | **Perplexity** | PerplexityBot, Perplexity-User | Search discovery and page retrieval for user requests. | **... and many more bots**, including those grouped under **Microsoft, Apple, Amazon, Meta, ByteDance, DuckDuckGo, Huawei, Common Crawl, and Internet Archive**. **Bot requests and shopper visits measure different activity.** One bot can make many requests. A request alone does not show that someone saw your brand in an AI answer, clicked through, or purchased. ## Turn the patterns into useful questions **Product schema** ### Which product pages should we prioritize for structured data? Review product URLs in **Most crawled paths** to find pages agents request frequently. For example, a popular juniper bonsai product page is a useful place to check whether its product schema accurately describes the item, price, currency, and availability. **Put it to work:** use crawl activity to prioritize a separate schema review. Agent Visitors records page requests; it does not validate schema or confirm which fields an agent extracted. **Collections vs. products** ### Are agents requesting collections or individual products? Compare paths such as **/collections/indoor-bonsai** and **/products/juniper-starter-kit**. Filter by platform or crawler to see which page types appear among its most requested URLs. **Put it to work:** for collection pages, review product groupings and links to individual items. For product pages, review variants, specifications, pricing, and availability. The table shows the top ten paths, so it is a starting point rather than a complete catalog breakdown. To explore where shoppers actually land, review AI-referred visits in Identity Mesh. **Catalog updates** ### Are agents revisiting pages after catalog changes? After adding a collection, changing availability, or updating product schema, compare requests before and after the change using the same bot or crawler selection. **Put it to work:** look for renewed activity on the affected product and collection URLs. A later request shows a visit after your change; it does not confirm that an assistant adopted the new price, availability, or schema. **New** means there were no recorded requests in the comparison period. The table's **First published** and **Most recently edited** dates refer to first and last observed bot requests. Compare these with the actual update dates in your catalog. ## Connect agent activity to shopper behavior Use Agent Visitors to understand automated requests, then explore human visits from AI assistants in [Human Visitors](/project/docs/bonsai-analytics/human-visitors.md). Where available signals allow, you can review what those visitors explored and whether they returned or purchased. > **[Identity Mesh](/project/docs/bonsai-analytics/identity-mesh.md)** > > Explore visits from AI assistants and the connected customer journeys available. > **[Analytics](/project/docs/bonsai-analytics/analytics.md)** > > Review shopper behavior, engagement, and commercial outcomes. This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.