Introduction

Erdo is a team of specialized AI agents that work on your business the way employees do — they connect to your data, answer questions, build things, and take action. You brief an agent in plain language; it does the work, shows you what it found, and asks before doing anything consequential.

​

How it works

You work with Erdo in a thread — a conversation with an agent. Each agent is specialized: one analyzes data and answers questions, another writes and runs code, another builds dashboards. You describe what you want; the agent connects to your data, does the work, and reports back with the result inline. Agents don’t start from scratch each time. They build up knowledge about your business — the metrics that matter, how your data is shaped, decisions you’ve made — and reuse it. Anything an agent proposes to remember or change, you can review first.

“Pull last month’s signups by plan, flag anything unusual, and build me a dashboard I can share with the team.”

That one request connects to your data, runs the analysis, and produces a live page — all in a single thread.

​

Beyond chat

The same agents work when you’re not watching:

• Automate — have an agent run on a schedule or a trigger, so the analysis (or the alert, or the report) happens on its own. See Automations.
• Build pages — turn a result into a shareable, data-wired page with live data and realtime state.
• Make calls — have an agent place an outbound phone call and book a meeting on your calendar.

​

Key concepts

​

Use Erdo anywhere

Erdo isn’t only the web app. Connect Claude, Cursor, or any MCP client to your data with the MCP server, or drive Erdo from the CLI — no code required, and right here under Product. Building software on Erdo? The REST API and the TypeScript / Python SDKs live in the Developers tab.