Getting Started

Follow these steps to go from a new account to your first dashboard.

Prerequisites

• A database you can connect to (PostgreSQL, Snowflake, BigQuery, MySQL, Redshift, DynamoDB).
• Network access for ContextFlo to reach your database (add the app’s IPs to your allowlist if needed).
• Credentials with read access to the schemas/tables you plan to query.

Step 0: Sign in with your provider to create an organization

1. Sign up with Google or Microsoft.
2. You’ll land on the home page with onboarding steps to guide setup.

Step 1: Connect a data source

1. Go to Data Sources → Add New.
2. Pick your database type.
3. Follow the instructions for each data source type to connect.

4. Click Test Connection. Fix any permission or network errors shown.
5. Click Save to store the connection.

Step 2: Sync tables

1. After saving, click on "Manage" to see list of tables.

2. Select the tables you want available in ContextFlo.
3. Click on Sync n Tables.

Step 3: Connect your context sources

1. Navigate to Context Sources → Add New Source.
2. Click on the relevant context source type (GitHub, GitLab, text input, etc.).

3. For GitHub/GitLab, it will take you through the installation process. For Text Input, you can save it instantly.

Step 4: Generate context

1. Navigate to Generate Context on the left sidebar under "Tasks"
2. Pick the tables and context sources. Make sure to pick the right table/context-source combination.
3. Click on "Generate Context" button.

4. It will take some time to generate context. Once context is generated, you can navigate to "Review Context" and check if the context is generated correctly for each table.

Step 5: Connect ContextFlo to the LLM

1. Pick your LLM client (e.g., Claude or ChatGPT Business).
2. Follow the corresponding guide to connect: Claude or ChatGPT Business.
3. Once connected, you can chat against your synced data and generated context.