Skip to content

Getting Started

Jump to bottom
Ömer Tarık Yılmaz edited this page Apr 5, 2026 · 1 revision

Getting Started

Welcome to White-Ops, a multi-agent task automation platform. This guide will take you from zero to running your first autonomous agent in under five minutes.

Table of Contents

Prerequisites

Requirement Minimum Recommended
Docker 24.0+ Latest
Docker Compose v2.20+ Latest
RAM 4 GB 8 GB+
Disk 5 GB 20 GB+
OS Linux, macOS, Windows (WSL2) Ubuntu 22.04+ / macOS 13+

You also need at least one LLM provider configured:

Provider Key Required Notes
Anthropic (Claude) ANTHROPIC_API_KEY Recommended default
OpenAI (GPT-4o) OPENAI_API_KEY
Google (Gemini) GOOGLE_API_KEY
Ollama None Local models, no API key needed

5-Minute Setup

1. Clone the Repository

git clone https://github.com/your-org/white-ops.git
cd white-ops

2. Create Your Environment File

cp .env.example .env

Open .env and set the required secrets:

# Generate secure random strings for these:
SECRET_KEY=your-random-string-at-least-32-characters
JWT_SECRET_KEY=another-random-string-at-least-32-characters

# Set passwords:
POSTGRES_PASSWORD=a-strong-database-password
REDIS_PASSWORD=a-strong-redis-password
MINIO_ROOT_PASSWORD=a-strong-minio-password
ADMIN_PASSWORD=your-admin-login-password

# Configure at least one LLM provider:
ANTHROPIC_API_KEY=sk-ant-...

See Configuration for the full environment variable reference.

3. Start the Platform

make up

This pulls images, builds containers, and starts all services:

  • PostgreSQL -- persistent data store
  • Redis -- cache and message queue
  • MinIO -- S3-compatible file storage
  • Server -- API backend (port 8000)
  • Worker -- agent execution engine
  • Web -- admin panel (port 3000)
  • Mail -- internal agent mail server (port 8025)

4. Verify Everything Is Running

make status

All services should show Up (healthy). You can also check:

curl http://localhost:8000/health

5. Open the Admin Panel

Navigate to http://localhost:3000 in your browser.

Log in with the credentials from your .env:

  • Email: admin@whiteops.local (or your ADMIN_EMAIL)
  • Password: the value of ADMIN_PASSWORD

Your First Agent

  1. In the admin panel, go to Agents in the sidebar.
  2. Click Create Agent.
  3. Fill in the form:
    • Name: Research Assistant
    • Description: Searches the web and summarizes findings
    • LLM Provider: anthropic (or your configured provider)
    • LLM Model: claude-sonnet-4-20250514
    • Tools: Select web_search, summarizer, notes
  4. Click Save.
  5. Back on the Agents list, click the Start button next to your agent.

Alternatively, use one of the 12 built-in Admin-Panel-Guide#agent-presets for a pre-configured profile.

Your First Task

  1. Go to Tasks in the sidebar.
  2. Click Create Task.
  3. Fill in the form:
    • Title: Research AI Trends 2025
    • Description: Search for the top 5 AI trends in 2025. Summarize each in 2-3 sentences and save the results as a note.
    • Agent: Research Assistant
    • Priority: medium
  4. Click Create.

The task is now queued. The worker will pick it up, your agent's LLM will plan and execute using the assigned tools, and results will appear on the task detail page.

Monitor progress in real time via the Dashboard or the Activity Feed.

Demo Mode

To populate the platform with sample data for exploration:

make seed

This creates:

  • 5 sample agents with different tool configurations
  • 10 sample tasks in various states (pending, in_progress, completed, failed)
  • Sample workflows and message threads
  • Example files and knowledge base entries

Demo mode is ideal for evaluating the admin panel and understanding the platform before creating real agents.

Next Steps

Topic Page
Understand the system design Architecture
Explore the admin panel Admin-Panel-Guide
Browse the 55 available tools Tool-Reference
Build multi-step workflows Workflow-Builder
Add more worker machines Worker-Management
Configure LLM providers and settings Configuration
Deploy to production Deployment-Guide
Secure your installation Security
Integrate with the API API-Reference
Contribute to the project Contributing
Common questions FAQ

Clone this wiki locally