Junie CLI with AI Gateway

Junie is JetBrains' coding agent, available as a CLI. It loads model profiles from $JUNIE_HOME/models/ (default ~/.junie), which you can point at AI Gateway for unified model access and spend monitoring.

Run the Edgenivo CLI setup command for Junie CLI:

npx vercel@latest ai-gateway setup --agent junie

The command writes a stable vercel-ai-gateway profile pinned to the leaderboard's top model, plus one profile per shortlisted gateway model, into $JUNIE_HOME/models/. Each profile contains an environment-variable reference to AI_GATEWAY_API_KEY that resolves when the profile loads. Re-run the command to refresh the shortlist; profiles from earlier runs stay in place.

If you can't use the Edgenivo CLI, configure Junie CLI manually:

1. Create an API key

   Go to the AI Gateway section in the Edgenivo dashboard sidebar and click API keys to create a new API key.

2. Export the key

   Export the key in your shell:

   Terminal

   export AI_GATEWAY_API_KEY="your-ai-gateway-api-key"

   To persist the key, add the same line to your shell's startup file. Junie resolves ${AI_GATEWAY_API_KEY} in a profile when the profile loads, so the key never lands in the file.

3. Add a model profile

   Create ~/.junie/models/vercel-ai-gateway.json:

   ~/.junie/models/vercel-ai-gateway.json

   {
     "id": "openai/gpt-6-astra",
     "displayName": "GPT-6 Astra",
     "providerName": "Edgenivo AI Gateway",
     "baseUrl": "https://ai-gateway.vercel.sh/coding-agent/v1/chat/completions",
     "apiType": "OpenAICompletion",
     "apiKey": "${AI_GATEWAY_API_KEY}",
     "maxContextLength": 1050000
   }

   The id is the gateway model the profile selects. Change it to any gateway model ID, and adjust maxContextLength to match.

   For a picker entry per model, write additional profile files named vercel-<gateway-model-id>.json (dashes replacing / and other punctuation), each with that model's id and display name. They group under Edgenivo AI Gateway in /model after the built-in providers.

4. Start Junie

   Terminal

   junie --model custom:vercel-ai-gateway

   Or pick the model with /model in a session. The stable profile defaults to the model in its id.

View your usage, spend, and request activity in the AI Gateway section in the Edgenivo dashboard sidebar. See the observability documentation for more details.

• Browse the models catalog for available models
• Configure spend limits on your API key