Roo Code
Wire: OpenAI. Base URL: https://sator-api.princep.org/v1. State: partially verified — the fields are Roo Code's documented ones; the /v1 form of the base URL is inferred from the reference URL in its docs rather than stated there.
Install#
Install the Roo Code extension in VS Code from the marketplace.
Configure#
In Roo Code's settings, add a configuration profile:
| Field | Value |
|---|---|
| API Provider | OpenAI Compatible |
| Base URL | https://sator-api.princep.org/v1 |
| API Key | your sk-sator-v1-… key |
| Model | an id from Models, for example deepseek-v4-flash |
| Context Window | the model's context window, e.g. 1000000 |
| Max Output Tokens | the model's max output, e.g. 384000 |
| Image Support | tick only if you send images |
Roo Code uses native tool calling exclusively. Every Sator model supports tools on this wire, and streamed tool-call deltas are forwarded exactly as the model emits them. See Tool calling.
Verify#
Ask for something trivial: Reply with exactly: ok. A working setup answers ok; the request appears in your dashboard.
Troubleshooting#
- 404 on every request — the base URL is missing its
/v1. - Model list is empty — the base URL is wrong; the list is filled from
{base URL}/models, which works without a key. - "Model does not support tool calling" — the model id is not a catalog id, or the request never reached Sator. Check
GET /v1/models.