Quick Start

Get Running in Under 5 Minutes

Choose the setup method that works best for you. Visual Studio gives you the full development experience with debugging; Docker is the fastest way to try OpenCaddis out.

Prerequisites
  • Visual Studio 2022+ (17.x) or VS Code with C# Dev Kit
  • .NET 10 SDK
  • Git

Option 1: Visual Studio (Recommended)

The recommended approach for development. Gives you full debugging, hot reload, and IntelliSense.

Step 1: Clone the repository
git clone https://github.com/vulcan365/OpenCaddis.git
Step 2: Open in Visual Studio

Open OpenCaddis.sln in Visual Studio. Set OpenCaddis as the startup project.

Step 3: Configure

Copy the example config files and add your LLM provider details:

# In the src/OpenCaddis directory:
copy opencaddis.json.example opencaddis.json
# Edit fabr.json — add your Azure OpenAI or OpenAI API key and endpoint
# Edit opencaddis.json — configure your agents and plugins
Step 4: Run

Press F5 (or Ctrl+F5 without debugger). Your browser opens to http://localhost:5000.


Option 2: Docker

The fastest way to try OpenCaddis. Available on Docker Hub.

# Pull and run
docker pull vulcan365/opencaddis:latest
docker run -d -p 5000:5000 --name opencaddis vulcan365/opencaddis:latest

Then open http://localhost:5000 in your browser.

Docker Limitation

When running in Docker, plugins like PowerShell, FileSystem, and Docker operate inside the container — they do not have access to your host machine's files, processes, or Docker engine. For full plugin capabilities (host file access, PowerShell on your machine, host Docker control), run from Visual Studio or the .NET CLI instead.

Useful Docker commands
docker stop opencaddis       # Stop the container
docker start opencaddis      # Start it again
docker logs opencaddis       # View logs
docker rm opencaddis         # Remove the container

Option 3: .NET CLI

git clone https://github.com/vulcan365/OpenCaddis.git
cd OpenCaddis/src/OpenCaddis
dotnet run

Then open http://localhost:5000 in your browser.


First Steps After Launch

Once running, configure your LLM provider in fabr.json and create agents in opencaddis.json. See the Configuration guide for full details.

Explore the Documentation

Documentation