Connecting a database gives Genloop direct access to your structured data. It reads from your source without creating copies or requiring data migrations. Supported databases: MySQL, PostgreSQL, Snowflake, BigQuery, Redshift, Databricks. From the Connections page, click Connect database under the Structured Data tab to start.Documentation Index
Fetch the complete documentation index at: https://docs.genloop.ai/llms.txt
Use this file to discover all available pages before exploring further.
The 4-step connection flow
Step 1 — Select your database Choose your database type from the list of supported sources. [Screenshot: Database type selection] Step 2 — Add credentials Enter your connection details: host, port, database name, and credentials.The credentials you provide need read access to the tables you want Genloop to work with. Write access is not required. Genloop connects directly to your source — no data copies are created.
.md, .txt, .pdf, or .docx format. This helps Genloop understand not just the shape of your data, but how your business uses it.
[Screenshot: Table and column selection with context document upload]
Step 4 — Memory is created Genloop processes your selected tables and builds your workspace context. You will see a progress stream as it runs. If there is an error at any point, you can retry without starting over.
[Screenshot: Memory generation loading stream]
Once complete, your connection is live and ready to query in Chat.
Managing existing connections
Your connected databases are listed in the Structured Data tab with the source name, connection date, and last synced time. You can edit or remove a connection from the actions menu on each row.Next steps
Upload a data file
Add CSV or XLSX files alongside your database connection.
Context Hub
View and edit the memory Genloop built from your connection.