Absolutely valid concern. Our extension connects to LLMs through Github Copilot. Github Copilot is Microsoft product and offers variety of enterprise plans, which enables your IT to approve what can be used for what kind of data. This gives you a clear path towards compliance with your enterprise requirements.
Isn't it a problem with many-to-many relationships in relational algebra? You have to have this artificial table to relate things in many-to-many way. How would you display it in a better way?
PM on the project here - The results from the query are generally not used by the LLM. In agent mode though, during query planning, the agent may retrieve sample of the data to improve precision of the queries. For example, getting distinct values from dimensional table to resolve filter condition from natural language statement.
Not directly in the extension connection object, but it's in on the roadmap. In the meantime, you can create ssh tunnel using vscode remotes and then just connect through it.
PM on the project here - All clouds and on-prem/local/docker work great! We test those, so if you have issues, please report in the repo we'll fix them.
I am PM from the project - short answer is yes - you can connect to any Postgres endpoint. There are some Azure specific features, like EntraID supported by the extension that don't exist in other cloud providers.