> For the complete documentation index, see [llms.txt](https://docs.airops.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.airops.com/insights/onsite/traffic.md).

# Agent Analytics

See how AI crawlers and assistants access your site

Open **Insights → Onsite → Agent Analytics** to see how AI systems reach your site: which engines crawl your content, whether they're crawling for search or training, and when an AI assistant sends a real person to a page. Agent Analytics draws on your CDN request logs, so there's no analytics script to install.

Agent Analytics must be enabled for your workspace, and you need to connect a supported CDN or hosting provider to collect data. If the tab is unavailable, contact your AirOps team to check access.

## How to Configure

1. Choose the guide for your provider:
   * [Cloudflare Logpush](/insights/onsite/traffic/cloudflare-logpush-agent-analytics.md)
   * [Cloudflare Worker](/insights/onsite/traffic/cloudflare-worker-agent-analytics.md)
   * [CloudFront](/insights/onsite/traffic/cloudfront-agent-analytics.md)
   * [Netlify](/insights/onsite/traffic/netlify-agent-analytics.md)
   * [Vercel](/insights/onsite/traffic/vercel-agent-analytics.md)
2. Complete the connection steps in the provider guide.
3. Return to **Insights → Onsite → Agent Analytics** to review data as requests arrive.

## Core Metrics

| Metric            | What it counts                                                                                                                                                                                     |
| ----------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| Total Crawls      | Every automated AI crawl of your site during the selected period (web search and training crawls combined). A higher number means AI systems are fetching more of your content.                    |
| Web Search Crawls | Crawls from AI search engines fetching a page to ground a live answer, such as ChatGPT search and Perplexity. These track content being pulled into real-time AI answers.                          |
| Training Crawls   | Crawls that collect your content to train AI models, such as GPTBot and ClaudeBot.                                                                                                                 |
| AI Agent Visits   | Visits from an AI assistant's browser acting on behalf of a person, for example when someone in ChatGPT or Claude follows a link to your site. Each visit is one request, counted when it happens. |
| Top Engine        | The AI engine that crawled your site the most during the selected period.                                                                                                                          |

## Charts

| Chart                              | What it shows                                                                                                                            |
| ---------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------- |
| Crawl activity vs. AI agent visits | How crawling and AI agent visits move over time, so you can connect a rise in crawls to the visits that follow.                          |
| Engine breakdown                   | Training, web search, total crawls, and AI agent visits for each AI engine, so you can see which platforms are most active on your site. |

## Per-page metrics

These metrics are also available for each URL in the [Pages](/insights/onsite/pages.md) view. Every page row shows its crawl and AI-agent-visit counts, and opening a page shows the same metrics and trends for that single URL.

## Query Agent Analytics through MCP or the API

Use AirOps MCP in a connected assistant or Playbook to compare page-level AI Agent Visits, Training Crawls, and Web Search Crawls with citation data. Agent Analytics must already be connected to collect these metrics, and MCP resource access requires a paid workspace, including an eligible paid-tier trial.

The `list_pages` tool exposes those three counts. Use `list_aeo_agent_analytics` for daily rows by engine and crawl type, then aggregate all pages of results for site totals and trends. See the [MCP tools reference](/developers/tools.md#list_aeo_agent_analytics) and [API calculation guide](/developers/api.md#agent-analytics).

For example, ask: "For the past 30 days, which tracked pages received AI agent visits but few citations?" Crawl activity records requests to your pages; citations measure whether those pages appear in tracked answers.

## Filters

* **Date**: Select a date range for your analysis
* **Data sources**: Choose which connected CDNs to include


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.airops.com/insights/onsite/traffic.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.