> For the complete documentation index, see [llms.txt](https://docs.datasentinel.io/manual/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.datasentinel.io/manual/ai-connector/analyze-and-tune-workloads/browse-your-fleet.md).

# Browse and navigate your fleet

Display your monitored PostgreSQL inventory at a glance, then narrow it down by tag until you reach the instance you want to work on.

Before analyzing anything, you often want to know **what is monitored**: how many instances, in which environments, for which applications, in which datacenters. Ask the assistant for your inventory and it answers with an overview of the fleet, then lets you narrow it down by tag until you reach the instance you want to work on.

**Example prompts:**

* "Summarize my production PostgreSQL inventory."
* "Which instances have not reported any metrics in the last hour?"
* "List every instance of zabbix application with its version and size."
* "Show me the instances in the dc1 datacenter."

## What the inventory shows

The assistant starts with a glance-level overview: the total number of monitored instances, plus the three [tag (TAG)](https://app.gitbook.com/o/DaDQkGCM7r2TghFSBFLQ/s/yE16Xb3IemPxJWydtPOj/features/tips-and-hints/tags) axes with the most variety in your fleet (environment, application, datacenter, and so on), each with its distinct values.

The overview below is an example. The axes and values you get depend on the tags applied to your own instances: a fleet tagged by application and location shows those axes instead.

| Axis         | Value                                                         |
| ------------ | ------------------------------------------------------------- |
| Instances    | 42 monitored                                                  |
| Applications | 7 (crm, billing, analytics, search, identity, audit, support) |
| Environments | 3 (production, staging, development)                          |
| Datacenters  | 4 (eu-west, us-east, ap-south, eu-north)                      |

Each instance then comes with the details you need to pick it out: name in `host@instance` form, server and port, role (primary or replica), PostgreSQL version, size, databases, and the time of its last collection. A last collection far behind the current time means monitoring has stopped for that instance, and the assistant points it out.

{% @arcade/embed url="<https://app.arcade.software/share/pObmcT2CAWFM5n4BXKfr>" flowId="pObmcT2CAWFM5n4BXKfr" %}

*This walkthrough was recorded with the Claude Desktop app.*

## Navigate from the fleet to one instance

On a large fleet, the assistant does not dump every instance at once. When the list is long, it asks you to pick a value on the tag axis that best partitions it (an environment, then an application, for example) and filters step by step until the list is short enough to read.

You can also narrow it yourself in the prompt: "the production instances", "the billing replicas", "everything in datacenter:Paris". Once you name an instance, the assistant keeps it as the target for the rest of the conversation, so the following analyses apply to it without asking again.

{% hint style="info" %}
Every inventory ends with a short menu of next steps tied to the instances shown: rank them by load, drill into one, or check the ones that stopped reporting.
{% endhint %}

{% hint style="warning" %}
On a large fleet, the quality of this navigation depends directly on the quality of your tagging. An instance with a missing or misspelled tag silently drops out of every filter that selects by that tag. See [What you can analyze](/manual/ai-connector/analyze-and-tune-workloads/what-you-can-analyze.md#on-large-fleets-tag-your-instances).
{% endhint %}

Once you have located a group of instances, run a cross-instance analysis on it: see [Analyze across multiple instances](/manual/ai-connector/analyze-and-tune-workloads/analyze-across-instances.md).


---

# 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.datasentinel.io/manual/ai-connector/analyze-and-tune-workloads/browse-your-fleet.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.