Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions docs.json
Original file line number Diff line number Diff line change
Expand Up @@ -112,6 +112,7 @@
"group": "AI agents",
"icon": "robot",
"pages": [
"guides/ai-overview",
"guides/ai-agents",
"guides/ai-agents/getting-started",
"guides/ai-agents/using-ai-agents",
Expand Down
154 changes: 154 additions & 0 deletions guides/ai-overview.mdx
Original file line number Diff line number Diff line change
@@ -0,0 +1,154 @@
---
title: "Using AI with Lightdash"
sidebarTitle: "AI overview"
description: "Lightdash offers multiple AI-powered capabilities to help you explore data, build your semantic layer, and integrate with external tools."
---

Lightdash provides several ways to leverage AI across your analytics workflow. Each capability serves a different use case—from answering business questions to building your semantic layer to integrating with your own AI tools.

## AI capabilities at a glance

| Capability | Use case | Requirements |
|------------|----------|--------------|
| [Lightdash AI agents](#ai-agents) | Answer questions and create charts using your semantic layer | Cloud Pro or Enterprise add-on |
| [Lightdash Data MCP](#lightdash-data-mcp) | Query your semantic layer from external LLMs | Cloud Pro or Enterprise add-on |
| [Lightdash Docs MCP](#lightdash-docs-mcp) | Give your LLM context about Lightdash features | Free for everyone |
| [Agent skills for your CLI](#agent-skills) | Build and maintain your semantic layer with external AI coding agents | Free for everyone |

## Lightdash AI agents

<Info>
Available as an add-on for Cloud Pro and Enterprise plans. [View pricing](https://www.lightdash.com/pricing)
</Info>

AI agents let your team ask questions in natural language and get instant answers. They use your saved dashboards and the metrics defined in your semantic layer, so business users can find insights without writing code or learning complex tools.

**Best for:** Business users who want to explore data and create visualizations without technical knowledge.

**Key features:**
- Natural language queries with automatic visualizations
- Slack integration for team collaboration
- Memory and learning from corrections
- Fine-grained access control
- Evaluations to validate agent accuracy

<Card title="AI agents documentation" icon="robot" href="/guides/ai-agents">
Learn how to set up and use AI agents
</Card>

## Lightdash Data MCP

<Info>
Available for Cloud Pro and Enterprise plans. [View pricing](https://www.lightdash.com/pricing)
</Info>

The [Model Context Protocol (MCP)](https://modelcontextprotocol.io/) lets you connect external AI assistants like Claude or ChatGPT directly to your Lightdash data. Use your preferred LLM to explore data models, search for metrics, and get data-driven insights—all while respecting your existing access controls.

**Best for:** Teams who want to use their own AI tools (Claude, ChatGPT, custom agents) to query Lightdash data outside the Lightdash UI.

**Key features:**
- Connect Claude, ChatGPT, or custom agents to your Lightdash instance
- Browse and query your data models
- OAuth authentication with existing permissions
- Read-only access for security

<Card title="Lightdash Data MCP documentation" icon="plug" href="/references/integrations/lightdash-mcp">
Set up MCP with your AI assistant
</Card>

## Lightdash Docs MCP

<Check>
Free for everyone—open source users and all cloud tiers.
</Check>

The Docs MCP endpoint gives AI coding agents access to the complete Lightdash documentation. This helps your LLM understand Lightdash concepts, generate accurate YAML configurations, and troubleshoot issues using official documentation.

**Best for:** Developers who want their AI coding tools to have context about Lightdash when working on configurations.

**Key features:**
- Full access to Lightdash documentation
- No authentication required
- Works with Cursor, Claude Code, Windsurf, and other MCP-compatible tools

### Setup

Add the Docs MCP endpoint to your AI coding tool:

<Tabs>
<Tab title="Cursor">
Add to your `.cursor/mcp.json`:

```json
{
"mcpServers": {
"lightdash-docs": {
"url": "https://docs.lightdash.com/mcp"
}
}
}
```
</Tab>
<Tab title="Claude Code">
Run the following command:

```bash
claude mcp add lightdash-docs https://docs.lightdash.com/mcp -t http
```
</Tab>
<Tab title="Windsurf">
Add to your `~/.codeium/windsurf/mcp_config.json`:

```json
{
"mcpServers": {
"lightdash-docs": {
"serverUrl": "https://docs.lightdash.com/mcp"
}
}
}
```
</Tab>
</Tabs>

## Agent skills

<Check>
Free for everyone—open source users and all cloud tiers.
</Check>

Agent skills teach AI coding agents like Cursor, Claude Code, and Codex about Lightdash's YAML configuration, metrics, dimensions, and best practices. Install skills to help your AI assistant generate correct configurations and follow Lightdash conventions.

**Best for:** Developers building and maintaining their semantic layer who want AI assistance with YAML configurations.

**Key features:**
- Generate correct YAML for metrics and dimensions
- Follow Lightdash naming conventions automatically
- Understand table relationships and joins
- Stay current with latest Lightdash features

### Quick start

```bash
# Install for Claude (default)
lightdash install-skills

# Install for Cursor
lightdash install-skills --agent cursor

# Install for Codex
lightdash install-skills --agent codex
```

<Card title="Agent skills documentation" icon="wand-magic-sparkles" href="/guides/developer/agent-skills">
Learn more about installing and using agent skills
</Card>

## Choosing the right AI capability

| If you want to... | Use |
|-------------------|-----|
| Let business users ask questions in natural language | [Lightdash AI agents](/guides/ai-agents) |
| Query Lightdash data from Claude, ChatGPT, or custom agents | [Lightdash Data MCP](/references/integrations/lightdash-mcp) |
| Help your coding AI understand Lightdash documentation | [Lightdash Docs MCP](#lightdash-docs-mcp) |
| Get AI help writing YAML configurations | [Agent skills for your CLI](/guides/developer/agent-skills) |