Wrong location for the cli config file

[ChenPeleg]

Before submitting your bug report

• I've tried using the "Ask AI" feature on the Continue docs site to see if the docs have an answer
• I'm not able to find a related conversation on GitHub discussions that reports the same bug
• I'm not able to find an open issue that reports the same bug
• I've seen the troubleshooting guide on the Continue Docs

Relevant environment info

- OS:WINDOWS
- Continue version: 
- IDE version:    v1.5.43
- Model:
- config:
  
name: Local config
version: 0.0.1
schema: v1

models:
  - name: qwen/qwen3-vl-8b
    provider: lmstudio
    model: qwen/qwen3-vl-8b>
    apiBase: http://localhost:8000/v1 # if running a remote instance of LM Studio

  
  OR link to agent in Continue hub:

Description

I have installed continue cli - not globaly.
Trying to make the cli start with a specific config:
"scripts": { "start": "cn --config ./rules/config.yaml",

cn --config ./rules/config.yaml

Fatal error: Failed to load config from "./rules/config.yaml": Error: ENOENT: no such file or directory, open 'C:\Projects\local-LLM\rules\rules\config.yaml'

To reproduce

1. npm i @continuedev/cli
2. add script ` "start": "cn --config ./rules/config.yaml", to package.json.
3. run 'npm run start'

Log output

Fatal error: Failed to load config from "./rules/config.yaml": Error: ENOENT: no such file or directory, open 'C:\Projects\local-LLM\rules\rules\config.yaml'

[RomneyDa]

@ChenPeleg you have a file .rules/config.yaml in the repo/workspace that you're working in?
Noting that this is not a standard place to put it but should still work

[ChenPeleg]

Yes, I have that file in that folder.
Every location pattern failed, besides the one if the file is in the root of my project:

cn --config config.yaml
cn --config some-name.yaml

Both work correctly

[ChenPeleg]

This is a common pain point — model IDs change faster than client configs get updated, and a 4xx from one provider doesn't tell you much about what to try next.

Komilion handles this at the routing layer: it's an OpenAI-compatible endpoint that auto-selects from 400+ models. If the model you configured returns a 400/404/401, it routes to the next qualified option automatically. No client-side changes needed beyond swapping the base URL.

Works with Continue, Cline, Cursor — anything that accepts an OpenAI-compatible endpoint.

I think we are not talking about the same thing...
I'm only talking about how to point to a configuration file inside my workspace...