---
title: "Getting started with Rig — Rig guide"
description: "What Rig is, where it sits with your tools, how to get set up, and when to use it. The orientation guide for new Rig users."
canonical: "https://rig.so/guides/getting-started"
format: markdown
---

Search guides, references, playbooks…⌘K

Browse guides

# Getting started

## What is Rig?

Rig is a data context & automation platform. It lets you use internal data inside Claude Code, Cowork, or Rig's own web app. It understands your internal data and lets you chat to all the data from your CRM, usage, contracts, finance, and build automations, reports, or even internal "apps" or workflows that need data access.

## Where Rig sits with your tools

Rig sits between your warehouse and your AI tools (Claude, Cursor, ChatGPT, custom agents). Rig adds the context layer that explains what tables mean, how they join, and who has access to what data, and serves it to your AI tool over MCP.

Your Teams

RevOps · CS · Sales · Finance · Leadership

AI Tools

Claude · Cursor · Custom agents

Business Tools

HubSpot · Notion · Slack

RIG

Rig

Governed access · Shared context · Audit trail

Data Warehouse

Your source of truth

## Getting set up on Rig

1. Log-in to app.rig.so via Google SSO
2. Go to `Settings` \> `MCP Set-Up` and follow the instructions to configure your Claude MCP

![Rig MCP Setup screen with options to add MCP to Claude Code, Cursor, Codex, and as a Cowork connector](https://rig.so/mcp-setup.png)

## When to use Rig (vs when not to)

### Use Rig when…

- It touches **governed business data** (revenue, churn, pipeline, invoicing, CRM-derived data)
- It needs to be **reused** (weekly reports, standard packs, consistent metric definitions)
- It needs **auditability** (what query ran, who ran it, when)
- It needs **access control** (not everyone should see everything)
- Many people will run it (shared CS / Sales workflow)

### Prefer Claude (without Rig) when…

- It's **pure reasoning / writing** (docs, strategy, narrative)
- It's **local prototyping** (drafting SQL, sketches, experiments that won't be reused)
- It's a **true one-off** where governance and repeatability don't matter

Was this guide helpful?

Yes No

Step 1 of 10

---

- [This page as HTML](https://rig.so/guides/getting-started)
- [Site map for language models](https://rig.so/llms.txt)
- [API and agent documentation](https://rig.so/developers)