> 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/use-the-assistant/explore-the-documentation.md).

# Explore the documentation from your assistant

Give your assistant read access to the Datasentinel documentation through the MCP server that GitBook publishes for it, next to your platform.

The Datasentinel documentation at [docs.datasentinel.io](https://docs.datasentinel.io/manual) is published with GitBook, and every published GitBook site comes with its own MCP server. Add it to your assistant next to your Datasentinel platform, and the assistant can answer product questions from the documentation while it works on your databases: what a setting does, how to configure an alert, what an API route returns.

<figure><img src="https://196776047-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FDL4xMeOs1Z7fz2LBCNSv%2Fuploads%2Fgit-blob-587832defe1403e228d9e42a2208fad59b051015%2Fexplore-the-documentation.svg?alt=media" alt="Animated illustration with two example questions asked to an AI assistant connected to the Datasentinel documentation MCP server, named docs-datasentinel. Example 1: the user asks how to install Datasentinel. The assistant calls searchDocumentation, the pages Installation, Platform and Installation, Agent match, and it answers in three steps: install the platform on a dedicated Linux machine, enable your license key, then connect your PostgreSQL instances with one agent per server or agentless connections. Example 2: the user asks what the prerequisites are in PostgreSQL. The pages PostgreSQL clusters, Extensions and Monitoring User match, and the assistant answers: PostgreSQL 9.4 or later, the pg_stat_statements extension installed in the postgres database, a read-only monitoring user with pg_monitor, pg_read_all_settings and pg_read_all_stats, and the optional extensions pg_store_plans, system_stats and pg_buffercache. Each answer ends with links to the source pages."><figcaption><p>Two product questions, answered from the documentation with links to the pages.</p></figcaption></figure>

The two servers complement each other. The platform server, described in the rest of this section, reads your monitoring data and needs a sign-in. The documentation server reads the public manual, needs no sign-in, and never sees your data.

## The documentation MCP server

|                |                                                       |
| -------------- | ----------------------------------------------------- |
| URL            | `https://docs.datasentinel.io/manual/~gitbook/mcp`    |
| Transport      | Streamable HTTP (the only transport GitBook supports) |
| Authentication | None: the documentation is public                     |
| Access         | Read-only                                             |

{% hint style="warning" %}
**Do not name this server with the `datasentinel-` prefix.** The Datasentinel Skill treats every MCP server whose name starts with `datasentinel-` as a monitoring platform and expects the platform tools on it. Name the documentation server `docs-datasentinel`, as in the examples below, so `datasentinel platforms` keeps listing only your platforms.
{% endhint %}

## Add it to your assistant

The server takes only its URL: no token, no header, no sign-in step. Use the same command or file as for the platform server on your assistant's page, with the documentation URL and the `docs-datasentinel` name.

{% tabs %}
{% tab title="Claude Code" %}

```bash
claude mcp add --scope user --transport http docs-datasentinel https://docs.datasentinel.io/manual/~gitbook/mcp
```

{% endtab %}

{% tab title="Claude apps" %}
In the Claude desktop app or on claude.ai, open **Settings**, then **Connectors**, choose **Add custom connector**, and paste the URL `https://docs.datasentinel.io/manual/~gitbook/mcp`. No sign-in follows, since the server is public.
{% endtab %}

{% tab title="Codex CLI" %}

```bash
codex mcp add docs-datasentinel --url https://docs.datasentinel.io/manual/~gitbook/mcp
```

{% endtab %}

{% tab title="Copilot CLI" %}

```bash
copilot mcp add --transport http docs-datasentinel https://docs.datasentinel.io/manual/~gitbook/mcp
```

{% endtab %}

{% tab title="Antigravity CLI" %}

```bash
agy mcp add docs-datasentinel https://docs.datasentinel.io/manual/~gitbook/mcp
```

{% endtab %}

{% tab title="VS Code" %}
In `mcp.json`, under `servers`:

```json
{
  "servers": {
    "docs-datasentinel": {
      "type": "http",
      "url": "https://docs.datasentinel.io/manual/~gitbook/mcp"
    }
  }
}
```

{% endtab %}

{% tab title="Cursor" %}
In `.cursor/mcp.json` (or the global `~/.cursor/mcp.json`), under `mcpServers`:

```json
{
  "mcpServers": {
    "docs-datasentinel": {
      "url": "https://docs.datasentinel.io/manual/~gitbook/mcp"
    }
  }
}
```

{% endtab %}
{% endtabs %}

Any other MCP client works the same way: register a remote server of type HTTP with that URL. The certificate is a public one, so nothing is needed on the TLS side.

## Use it

Once the server is connected, the assistant consults the documentation whenever a question is about the product rather than about your data. Naming the documentation in the request makes it explicit:

```
Search the Datasentinel documentation: what does query-monitoring-limit do, and what happens when it is exceeded?
```

```
Using the documentation, list the alert check types available and their default thresholds.
```

The two servers combine naturally. After `datasentinel whoami` or an analysis, ask the assistant to explain a setting it mentioned, and it turns to the manual:

```
The last analysis mentioned autovacuum thresholds. Find in the Datasentinel documentation how the platform collects table statistics and how often.
```

Answers come with links to the pages on `docs.datasentinel.io`, so you can read the full text.

{% hint style="info" %}
The Datasentinel Skill drives the platform servers only; it does not call the documentation server. The assistant uses the documentation server on its own judgement, and always when you ask for it by name.
{% endhint %}

## Verify

Ask the assistant:

```
Search the Datasentinel documentation for the retention policy of metrics.
```

A working connection returns passages with links to `https://docs.datasentinel.io/manual/...` pages. If the assistant answers from its own knowledge without links, check that the server appears in its MCP list (for example `claude mcp list` or `/mcp`) with the exact URL above. Use the `/manual` URL as written: the site root `https://docs.datasentinel.io/~gitbook/mcp` only redirects to it, and not every MCP client follows redirects.


---

# 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/use-the-assistant/explore-the-documentation.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.