Junie CLI

Connect JetBrains Junie CLI to APIClaw with a custom model profile. Junie expects the full chat completions URL, not a bare /v1 base.

Prerequisites

Create an API key in the API Keys dashboard.
Copy a full model ID from the model hub.

Configure

bash

export APICLAW_API_KEY="sk-your-apiclaw-key"

The variable must be set in the same shell you run junie from, or the profile fails to load. Create a profile file at ~/.junie/models/apiclaw.json (or $JUNIE_HOME/models/ if you set that, or .junie/models/ per project):

json

{
  "id": "YOUR_MODEL_ID",
  "baseUrl": "https://apiclaw.biz/v1/chat/completions",
  "apiType": "OpenAICompletion",
  "apiKey": "${APICLAW_API_KEY}"
}

Setting	Value
`baseUrl`	`https://apiclaw.biz/v1/chat/completions`
`apiType`	`OpenAICompletion`
`apiKey`	Your APIClaw key or `${APICLAW_API_KEY}`

Run Junie with the custom profile:

bash

junie --model custom:apiclaw

(Use the filename without .json as the custom profile id.) You can also set JUNIE_MODEL, or switch with /model in interactive mode.

Verify

Send a short task, then confirm the request in logs.

Troubleshooting

Most OpenAI-compatible docs say "base URL without path". Junie is different: omit the full path and requests fail. For Responses API models, use .../v1/responses with apiType: "OpenAIResponses" instead.