Skip to main content

​
Step 0: Choose your operating mode

Before setup, decide which user experience you want:
  • Direct Analytics: fastest, minimal guardrails
  • Trusted Analytics V1: contracts + policies + bundles + semantic metrics
  • Trusted Analytics V2: V1 runtime with OpenMetadata-fed governance signals

Compare operating modes

Pick the right trust level for your users

​
Step 1: Install dazense-core package

pip install dazense-core

​
Step 2: Initialize a dazense project

dazense init
This command will ask you:
  • To name your project
  • If you want to connect a database (optional)
  • If you want to add a repo in agent context (optional)
  • To add an LLM key (optional)
  • If you want to setup a Slack connection (optional)
You can skip any optional question and configure them later in your dazense_config.yaml file.
This will create:
  • A new folder with your project name
  • An architecture for your context files
  • A dazense_config.yaml configuration file
  • A RULES.md file

​
Step 3: Verify your setup

cd to the project folder and run: 
dazense debug
This command checks your configuration and displays any issues.

​
Step 4: Synchronize your context

dazense sync
This will populate your context folder with your context files (data, metadata, repos, etc.).

​
Step 5 (recommended): Validate governance configuration

dazense validate
If you’re using Trusted Analytics V1/V2, this checks bundle/policy/semantic consistency before chat usage.

​
Step 6: Launch the chat and ask questions

You have two options to access the chat UI:

​
Option 1: Using dazense chat command

dazense chat
This will start the dazense chat UI. It will open the chat interface in your browser at http://localhost:5005.

​
Option 2: Using Docker

Instead of dazense chat, you can use Docker to run the UI: With built-in example: 
docker run -d \
  --name dazense \
  -p 5005:5005 \
  -e BETTER_AUTH_URL=http://localhost:5005 \
  metazense/dazense:latest
With your project: 
docker run -d \
  --name dazense \
  -p 5005:5005 \
  -e BETTER_AUTH_URL=http://localhost:5005 \
  -v /path/to/your/project:/app/project \
  -e DAZENSE_DEFAULT_PROJECT_PATH=/app/project \
  metazense/dazense:latest
Access the UI at http://localhost:5005 and add your LLM API key in the settings. From there, you can start asking questions to your agent!

​
Step 7: Evaluate your agent

First, create a folder tests/ with questions and expected SQL in yaml. Then, measure agent’s performance on examples with dazense test command: 
dazense test
View results in tests panel: 
dazense test server
If you’re using Trusted Analytics V1/V2, also run: 
dazense eval
dazense eval --engine
These commands validate governance behavior, not only answer correctness.

Evaluation Guide

Learn how to build comprehensive test suites and evaluate your agent

​
What’s Next?

Context Builder

Learn how to build and customize your agent’s context

Self-Hosting

Deploy your agent in production

Chat Interface

Explore the chat interface features

dazense Cloud

Use our managed cloud service