Overview
This guide walks you through deploying Manifold locally using Docker. Follow these steps to get up and running quickly.
Prerequisites
- Docker installed and running
- Git for cloning and updating submodules
- An OpenAI API key (for AI features)
Step 1: Prepare Environment Files
Rename the example environment and config files:
cp example.env .env
cp config.yaml.example config.yaml
Step 2: Configure OpenAI API Key
Edit your .env file to set your real OpenAI API key:
# Replace test123 with your actual key
sed -i '' 's/^OPENAI_API_KEY="[^"]*"/OPENAI_API_KEY="your-key-here"/' .env
Step 3: Update Submodules
Initialize and update git submodules:
git submodule update --init --recursive
Step 4: Create Log File
Create a log file for Manifold:
touch manifold.log
Step 5: (Optional) Enable Web Search
To enable web search and fetch without SearXNG, pull the DuckDuckGo container:
docker pull mcp/duckduckgo
Step 6: Deploy Manifold
Start the minimal deployment (this may take a few minutes):
docker compose up -d manifold pg-manifold
Access the Web UI
Once containers are running, open your browser and go to:
Troubleshooting
- Check
manifold.logfor errors. - The most common issue is database DSN misconfiguration. Ensure your
.envandconfig.yamlmatch the Docker Compose settings for thepg-manifoldservice.
For more details, see the Manifold Documentation or contact the development team.