> ## Documentation Index
> Fetch the complete documentation index at: https://docs.citino.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Integrations

> Connect Google Analytics 4, Slack, CMS publishing to your WordPress or Webflow site, and your AI crawler logs to a project.

**Integrations** is per project: open the project and choose **Integrations** in the rail. The screen sums itself up as "What Citino reads, what it writes into, and what it cannot have."

| Integration | What it is for | Who can connect it |
| - | - | - |
| Google Analytics 4 | Visits AI engines send to your site, for [Attribution](https://docs.citino.ai/guides/screens/attribution.md) | Owner, admin, editor |
| Slack | The daily alert digest, posted to one channel | Owner, admin, editor |
| CMS publishing (WordPress or Webflow) | Sending an approved draft to your site, as a draft | Owner, admin |
| AI crawler logs | Which AI crawlers fetch your pages, for [Crawler activity](https://docs.citino.ai/guides/screens/crawler-activity.md) | Owner, admin, editor |

## Google Analytics 4

Connecting GA4 lets Citino match the visits AI engines send to your site back to the questions that produced them.

> Read-only. We request the Analytics read scope and nothing else, and you can disconnect from this page at any time.

**1. Connect**

On the Google Analytics 4 row, press **Connect**. Google's own consent screen opens.

**2. Allow read access**

Sign in to the Google account that can read the property, and allow access.

**3. Choose the property**

Back in Citino, choose which GA4 property to read. Each project reads one property.

Sessions arrive first and conversions fill in behind them. After that, a sync runs every night; starting one by hand is not available yet.

**What we import.** Totals per day, per referring site and per landing page: sessions, conversions and revenue. Never a person, a visitor id or any user-level data.

**Disconnecting.** Press **Disconnect and delete data**. "Imported sessions are deleted within 24 hours of disconnect."

**When Google stops the connection.** The row says the authorization expired. "Nothing already imported was deleted. Recent sessions are missing from Attribution until you reconnect, and we backfill what was missed when you do." Press **Reconnect**.

> [!NOTE]
>
> If somebody else owns the property, the row offers a setup request form. It does not send an email yet. Ask the property's owner to connect it, or invite them as an editor so they can.

## Slack

Slack posts this project's daily alert digest to one channel you pick on Slack's own screen.

1. Press **Add to Slack** and choose the channel.
2. Back in Citino, the row names the channel, and Slack is added to every rule on [Alerts](https://docs.citino.ai/guides/screens/alerts.md).

**Disconnect** takes Slack off every rule. If Slack stops accepting our messages, the row says so: press **Reconnect** and choose a channel again.

## CMS publishing (WordPress or Webflow)

This connection sends an approved draft from Gap detail to your site. As the screen says, it "Sends an approved draft to your WordPress or Webflow site as a draft. Nothing goes live until you publish it."

* **WordPress**: your site's address, a username and an application password. Citino checks them with WordPress before saving anything.
* **Webflow**: sign in with Webflow, then choose the collection drafts go to and its rich-text field.

> [!WARNING]
>
> "An application password can do anything your WordPress user can, on the whole site, and never expires. Delete it in WordPress (Users → Profile → Application Passwords) to revoke it."

A project connects one site at a time. To switch, disconnect the current one first. Disconnecting deletes the stored credential. Drafts are text only: add images in your CMS before you publish.

## AI crawler logs

This row shows which ways of sending logs are connected, and when one last sent something. Press **Set up** or **Manage** to open [Crawler activity](https://docs.citino.ai/guides/screens/crawler-activity.md), where setup lives. There are five ways in:

* **Cloudflare**
* **Vercel**
* **WordPress plugin**
* **Upload a log file**
* **Send from anything else**

We keep only requests from known AI crawlers to your own domain. The row warns when a connected way in has sent nothing for 48 hours. Your plan sets how many crawler visits we keep a month: see [Plans and prompt limits](https://docs.citino.ai/guides/account/plans-and-limits.md).

A crawler fetching a page is not the page being cited in an answer. Citino shows the two side by side and never adds them together.

## Related

> [card] [Attribution](https://docs.citino.ai/guides/screens/attribution.md)
>
> What GA4's visits turn into.

> [card] [Alerts](https://docs.citino.ai/guides/screens/alerts.md)
>
> The rules Slack's digest reports.

> [card] [Gaps](https://docs.citino.ai/guides/screens/gaps.md)
>
> Where an approved draft is sent from.

> [card] [Crawler activity](https://docs.citino.ai/guides/screens/crawler-activity.md)
>
> What the crawler logs show.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.