Cursor

1. Get a SupaNexus API key

Sign in to the Developer Console, open your project → API Keys, create a key, and copy the secret immediately (shown once). One project key works across OpenAI-compatible and Anthropic-compatible clients.

2. Open Cursor model / API settings

Open Cursor Settings (gear, or Cmd/Ctrl + ,) → Models. Expand the API Keys section. You will enable an OpenAI-compatible override rather than Cursor’s built-in vendor keys alone.

3. Enable OpenAI key + Base URL override

Toggle on OpenAI API Key and paste your SupaNexus project key. Toggle on Override OpenAI Base URL and set it to the SupaNexus OpenAI-compatible endpoint (host + /v1). Save / apply, then fully restart Cursor if the UI does not pick up the change.

OpenAI API Key: sk-snx-...
Override OpenAI Base URL: <BASE_URL>/v1

4. Add catalog models

In Models, click + Add model (or equivalent) and enter exact SupaNexus model IDs. Enable the models you added so they appear in the chat / agent picker.

Examples:
deepseek/deepseek-chat
anthropic/claude-3-5-sonnet
openai/gpt-4o

5. Pick a catalog model ID

Use exact IDs from the SupaNexus Models marketplace or GET /v1/models (for example deepseek/deepseek-chat, anthropic/claude-3-5-sonnet). Do not invent vendor-only slugs from other gateways.

6. Select the model and smoke-test

Open Chat or Agent, pick one of the models you added, and send a short prompt (e.g. explain a file). Confirm the reply lands and that usage appears under your SupaNexus project. Tab completion may still use Cursor’s built-in paths depending on plan — focus Agent / Chat on the override first.

7. Troubleshooting

401 / invalid key: confirm the key is a project API key and Base URL includes /v1. Model not found: the ID must exist in GET /v1/models. Tool / agent quirks: restart Cursor after changing Base URL; disable conflicting third-party OpenAI overrides.