Roo Code can use SoAI as an OpenAI-compatible model provider and as a separate Streamable HTTP MCP server.
Prepare SoAI
- In the SoAI WebUI, select the Settings button at the right of the header, choose Settings, and open
OpenAI. This tab appears only to an Administrator with API key management
(
OPENAI_API_ADMIN). Select Create key and copy the key when it is shown. Open Models and copy the exact model ID; this requires Model read access (MODEL_READ). - For MCP, select the Settings button at the right of the header, choose Settings, then open
MCP > MCP tokens. The MCP tab appears only to an Administrator with
MCP administration (
MCP_ADMIN), and creating a token also requires MCP usage (MCP_USE). Changing the MCP runtime, server mode, or exposed capabilities under MCP configuration also requires Configuration changes (CONFIG_PATCH).
Connect Roo Code to the SoAI model API
- Open the Roo Code pane, select its settings gear, and choose OpenAI Compatible as the API provider.
- Set the base URL to
https://SOAI_HOST:5090/v1, paste the SoAI API key, and enter an exact SoAI model ID. - Open the advanced model settings and declare the context window, output limit, image support, and other capabilities truthfully. Roo Code requires native tool calling for agent tool use.
- Save, select the model, and send a short task.
Connect Roo Code to SoAI MCP in its interface
- Select the MCP Servers icon in the top navigation of the Roo Code pane.
- Scroll down and select Edit Global MCP, or Edit Project MCP for a workspace configuration.
- Add the entry below, save it, and confirm that the server appears as connected.
- Leave
alwaysAllowempty initially and disable any tools the project does not need.
{
"mcpServers": {
"soai": {
"type": "streamable-http",
"url": "https://SOAI_HOST:5090/mcp",
"headers": { "Authorization": "Bearer YOUR_MCP_TOKEN" },
"alwaysAllow": [],
"disabled": false,
"timeout": 120
}
}
}