Just started exploring SuperAGI for a potential project at work. I was really excited after reading the intro docs—it seemed like the perfect toolkit to automate some of our reporting flows. But as soon as I went past the "Getting Started" page, I hit a wall.
The configuration examples in the docs keep referencing a `config.yaml` file structure that doesn't match what I see in the actual GitHub repo. I cloned the latest version, and the keys are completely different. For instance, the docs say to set `LLM_MODEL: gpt-4`, but the actual config file expects something like `model: "gpt-4"` under an `llm` section. Spent an hour trying to get an agent to run because of this. 😅
Also, the API endpoint examples for the local setup are off. The docs list ` http://localhost:8001` as the default, but the Docker setup seems to expose it on a different port. Had to check the Docker logs to figure it out. It feels like the documentation is for a version from several months ago.
Has anyone else run into this? I'm trying to build a simple agent that can connect to a PostgreSQL DB and run some basic analysis, but I'm stuck piecing together the correct setup from GitHub issues and forum snippets. Makes it really tough to evaluate if this is the right tool.