Skip to main content
This guide walks you through setting up Condor from scratch and building your first autonomous Condor Agent.

What You’ll Learn

1

Installing Condor and Hummingbot API

Deploy the two-server architecture on your machine
2

Telegram Interface

Control your trading infrastructure via mobile-friendly commands — or skip it entirely with Local and drive Condor from the browser
3

Workspace

Access the browser-based interface for managing bots and agents
4

Integrating your LLM

Connect Claude, Gemini, Copilot, Codex, OpenRouter, a local model, or any OpenAI-compatible endpoint for Condor Agents
5

Adding Credentials

Connect your exchange accounts via API keys
6

Seeing Your Portfolio

View balances and positions across all exchanges
7

Build Your First Agent

Create and deploy an autonomous Condor Agent
8

Agent Loops

Monitor, inspect, and control running agents

Prerequisites

Before starting, ensure you have:
  • A Linux server (Ubuntu 20.04+), a Mac, or Windows with WSL2 (Ubuntu)
  • 4 GB RAM minimum (8 GB recommended)
  • Exchange API keys with trade permissions
  • A Telegram account and bot token — unless you install in Local, which needs neither
  • Access to an LLM provider (Claude, Gemini, Copilot, Codex, OpenRouter, Ollama or LM Studio, or a custom OpenAI-compatible endpoint) for agent reasoning. See LLM Integration
  • Tailscale account (free tier is enough) — recommended when Condor and Hummingbot API run on different machines or any production VPS setup

Time to Complete

Most users complete this guide in 30 to 45 minutes.

Start Here

Begin with installation

Already Installed?

Jump to building your first agent
Already using Condor? Tell us what’s working and what you need next.

Help shape Condor

What’s working? What’s confusing? What do you wish Condor could do? Your answers go straight to the Hummingbot Foundation team and directly shape what we build next.