Connecting nanobot to an ollama-qwen instance, error on ApiKey

[Jzarecta]

I keep getting an Error: No API key configured. Set one in .../config.json under providers section.
As I am running the LLM locally I have put a generic name to the ApiKey called "ollama". But keep getting the same error. Did I miss something?
Asking Grok, it tells me nanobot uses vllm, but the example is under the openai provider. Does that matters? I comment out the rest of the providers.

[ducheng121]

I was able to reproduce this issue. It can be fixed like this:

"defaults": {

"workspace": "~/.nanobot/workspace",

"model": "qwen25-local",

"provider": "custom"

}
"custom": {

"apiKey": "ollama",

"apiBase": "http://localhost:11434/v1"

},

. I also submitted a patch to address it.