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

# Intelligence Overview

> AI that connects schema drift, freshness, data quality, and lineage into actionable answers

<div aria-hidden="true" style={{position:"absolute",width:"1px",height:"1px",overflow:"hidden",clip:"rect(0,0,0,0)",whiteSpace:"nowrap"}}>For LLM agents: documentation index at <a href="/llms.txt" tabIndex={-1}>/llms.txt</a>, full text at <a href="/llms-full.txt" tabIndex={-1}>/llms-full.txt</a>. Append .md to any page URL for plain markdown.</div>
Intelligence is the layer that ties your monitoring together. Schema drift detects changes. Freshness catches stale tables. Data quality tracks metrics. Intelligence reads all of that context and lets you ask questions, surface findings, and diagnose problems in plain English.

## How Intelligence Uses Your Monitoring Context

<Frame>
  <img src="https://mintcdn.com/anomalyarmor/CZXBGa_D1aE9spAI/images/diagrams/intelligence-context-flow-light.svg?fit=max&auto=format&n=CZXBGa_D1aE9spAI&q=85&s=e40d252f979d046220328659988cc177" alt="How monitoring features feed context into Intelligence for cross-domain diagnosis" className="block dark:hidden" width="800" height="380" data-path="images/diagrams/intelligence-context-flow-light.svg" />

  <img src="https://mintcdn.com/anomalyarmor/CZXBGa_D1aE9spAI/images/diagrams/intelligence-context-flow-dark.svg?fit=max&auto=format&n=CZXBGa_D1aE9spAI&q=85&s=5cea5c6b88e3d6c88506e707ea71ef4c" alt="How monitoring features feed context into Intelligence for cross-domain diagnosis" className="hidden dark:block" width="800" height="380" data-path="images/diagrams/intelligence-context-flow-dark.svg" />
</Frame>

Every AnomalyArmor feature generates context that Intelligence can reason over:

| Feature                                  | Context Created                                | What Intelligence Does With It                                            |
| ---------------------------------------- | ---------------------------------------------- | ------------------------------------------------------------------------- |
| [Schema Drift](/concepts/schema-changes) | Column additions, removals, type changes       | "A column was removed Tuesday. That's likely why your dashboard broke."   |
| [Freshness](/concepts/freshness)         | Update patterns, SLA violations, stale tables  | "The orders table is 6 hours overdue. Its upstream source is also stale." |
| [Data Quality](/concepts/metrics)        | Null rates, row counts, validity rule failures | "Null rate on email spiked from 2% to 28% after Wednesday's deploy."      |
| [Tags](/data-classification/overview)    | PII labels, domain classifications             | "3 tables with PII tags had schema changes this week."                    |
| [Lineage](/intelligence/overview)        | Upstream sources, downstream consumers         | "The stale staging table affects 7 downstream gold tables."               |
| [Alerts](/concepts/alerts)               | Alert history, rule configurations             | "You had 4 freshness alerts and 1 schema drift alert since Monday."       |

The more monitoring you set up, the smarter Intelligence gets.

## Features

<CardGroup cols={3}>
  <Card title="Object Profiles" icon="id-card" href="/intelligence/object-profiles">
    AI-generated profiles with findings, relationships, and change history for every table
  </Card>

  <Card title="Ask Questions" icon="message-question" href="/intelligence/ask-questions">
    Chat with an AI agent that has full context across all your monitoring domains
  </Card>

  <Card title="Walkthrough" icon="route" href="/intelligence/use-cases">
    See how monitoring context flows into Intelligence to solve real problems
  </Card>
</CardGroup>

## What You Can Ask

Intelligence answers are grounded in real data from your monitoring setup, not just schema names.

### Cross-Domain Diagnosis

```
"Why is my orders dashboard showing errors?"
"What changed upstream that could cause NULLs in silver_orders?"
"What should I prioritize this morning?"
```

### Schema and Structure

```
"What columns were removed this week?"
"Which tables have PII?"
"How is users related to orders?"
```

### Freshness and Health

```
"Which tables are stale right now?"
"What's the update pattern for my fact tables?"
"Show me SLA violations this week"
```

### Data Quality

```
"Which columns have spiking null rates?"
"What validity rules are failing?"
"Are there anomalies in my row counts?"
```

### Impact Analysis

```
"What's the blast radius if orders goes down?"
"What depends on the customers table?"
"Show me everything downstream of raw_events"
```

## The Intelligence Page

The Intelligence page is a two-panel layout:

* **Left**: Schema Explorer tree. Navigate databases, schemas, and tables.
* **Right**: Dynamic content that changes based on your selection.

Select a table to see its **Object Profile**: an AI-generated summary with findings grouped by severity, relationship mapping, change history, and which monitoring context is available. Select a database or schema to see a high-level overview with aggregated findings.

Click **Ask Agent** to open the full-screen chat, where you can ask questions scoped to the selected asset with full conversation history.

## Getting Started

<Steps>
  <Step title="Set up monitoring first">
    Intelligence is most useful when it has context to work with. Enable [freshness monitoring](/concepts/freshness), [schema drift detection](/concepts/schema-changes), and [data quality metrics](/concepts/metrics) on your key tables.
  </Step>

  <Step title="Generate intelligence">
    Navigate to a table on the Intelligence page and click **Analyze Table** (or **Analyze All** for the entire database). AI analyzes your schema and monitoring data to build a profile.
  </Step>

  <Step title="Ask questions">
    Click **Ask Agent** and ask about your data in plain English. The agent has access to all your monitoring context and can chain multiple tools to investigate issues.
  </Step>
</Steps>

## How It Works

Intelligence builds a knowledge base from your operational metadata:

| What We Analyze             | What We Never Access |
| --------------------------- | -------------------- |
| Table and column names      | Actual row data      |
| Data types and constraints  | Query results        |
| Schema change history       | Business values      |
| Freshness patterns and SLAs | Database credentials |
| Data quality metrics        |                      |
| Alert history and rules     |                      |
| Tags and classifications    |                      |
| Lineage and dependencies    |                      |

Your data values stay in your database. Intelligence only sees structure and operational metadata.

## Usage and Billing

Intelligence queries count against your monthly API limit:

| Plan         | Monthly Queries |
| ------------ | --------------- |
| Starter      | 500             |
| Growth       | 2,000           |
| Professional | 10,000          |
| Enterprise   | Unlimited       |

Check usage: **Settings** > **Billing** > **Usage**

## Privacy and Security

* Intelligence processes queries using AI language models (configurable per account)
* Only schema metadata and operational data are sent (names, types, alert history, freshness status)
* No actual data values are ever transmitted
* You can request Intelligence be disabled for your account

## Common Questions

### What does Intelligence actually do that monitoring alone doesn't?

Monitoring detects events, schema drift, stale tables, null rate spikes. Intelligence correlates them. It reads context from every monitoring domain and answers questions like "why is this broken?" in one place instead of you clicking through separate pages.

### Does Intelligence send my data to an LLM?

Only metadata: table names, column types, schema changes, freshness status, metric values, alert history, and tags. Row data, query results, and credentials are never transmitted. See the [Ask Questions](/intelligence/ask-questions) page for more on privacy.

### Do I need monitoring set up before Intelligence is useful?

Yes, effectively. Intelligence reasons over the context your monitoring creates. Without freshness schedules, drift detection, or metrics, it falls back to schema-only analysis and loses most of its diagnostic power.

### Is Intelligence usage capped?

Yes, by plan. Starter gets 500 queries per month, Growth 2,000, Professional 10,000, and Enterprise is unlimited. Check consumption under **Settings → Billing → Usage**.

### Can I turn Intelligence off for my account?

Yes. If your organization doesn't want AI processing, request that Intelligence be disabled. The rest of AnomalyArmor (monitoring, alerts, classification) continues working without it.

## Next Steps

<CardGroup cols={2}>
  <Card title="Object Profiles" icon="id-card" href="/intelligence/object-profiles">
    Understand what you see when you select a table
  </Card>

  <Card title="Ask Questions" icon="message-question" href="/intelligence/ask-questions">
    Chat with the AI agent
  </Card>
</CardGroup>
