Quickstart

Get KaaS running in under 3 minutes with a single Docker command. By the end of this guide you'll have submitted content, watched it compile into a structured wiki, and asked your first question.

Prerequisites

• Docker installed and running
• An OpenAI-compatible API key (OpenAI, DeepSeek, Ollama, etc.)

1. Start KaaS

docker run -d --name kaas \
  -p 8080:8080 \
  -v ./data:/app/data \
  -e LLM_API_KEY=sk-xxx \
  -e LLM_BASE_URL=https://api.openai.com/v1 \
  -e LLM_MODEL=gpt-4o-mini \
  kaas

Replace sk-xxx with your actual API key. If you use a different provider, update LLM_BASE_URL and LLM_MODEL accordingly.

Defaults

LLM_BASE_URL defaults to https://api.openai.com/v1 and LLM_MODEL defaults to gpt-4o-mini. You only need to set them if you're using a different provider.

2. Open the Web UI

Open your browser and navigate to:

http://localhost:8080

You should see the KaaS dashboard.

3. Submit Your First Content

1. Click Submit in the sidebar.
2. Paste some sample text — for example, a few paragraphs from a meeting note, a design doc, or any knowledge you'd like to turn into a wiki article.
3. Click Submit.

4. Watch It Compile

1. Navigate to the Status page.
2. You'll see your submission enter the compile pipeline: Extract → Classify → Write → Index.
3. Wait for the status to show Completed (usually 30–60 seconds depending on content length and your LLM provider).

5. Ask a Question

1. Click Chat in the sidebar.
2. Ask a question related to the content you just submitted — e.g., "What were the key decisions discussed?"
3. KaaS will retrieve the relevant compiled wiki articles and return an answer with source citations.

Congratulations — you've just experienced the full KaaS workflow: submit raw content, let it compile into structured knowledge, then query it conversationally.

What's Next

• Installation — full setup options (CLI install, building from source)
• Your First Knowledge Base — a detailed walkthrough of building a real knowledge base