Integrations

IDEs & Editors

Put the gateway between your IDE assistant and the model so every developer request is attributed, budgeted and auditable.

In every IDE below you are doing the same thing: use OpenAI (or OpenAI Compatible) as the provider, set the Base URL to the gateway, and use a gateway key as the API key.

1

Base URL: point your agent at the Spineforge gateway

https://spineforge-backend.onrender.com/gateway/v1
2

API key: use your Spineforge gateway key

sfgw-xxxx-xxxx
Prerequisite: add your actual LLM provider key (e.g., your Google or OpenAI key) to the Spineforge Dashboard Vault first, and create a gateway key in the API Keys tab. Developers only ever hold the sfgw- key, never the raw provider key.

Cursor

  1. Go to Settings > Models.
  2. Under OpenAI API Key, enter your Gateway Key (sfgw-...).
  3. Click Override OpenAI Base URL and enter https://spineforge-backend.onrender.com/gateway/v1.
  4. Add a model using a provider-prefixed name, for example openai/gpt-4o or gemini/gemini-3.5-flash. See Model Naming.

VS Code: native chat

  1. Press Cmd/Ctrl + Shift + P and select Chat: Manage Language Models.
  2. Click Add Models > Custom Endpoint.
  3. Set the group name to spineforge and paste your Gateway Key.
  4. Configure the model block in your settings.json:
[
  {
    "name": "spineforge",
    "vendor": "customendpoint",
    "apiKey": "${input:chat.lm.secret.-71f908a2}",
    "apiType": "chat-completions",
    "models": [
      {
        "id": "gemini/gemini-3.5-flash",
        "name": "Spineforge (Gemini 3.5 Flash)",
        "url": "https://spineforge-backend.onrender.com/gateway/v1",
        "toolCalling": true,
        "vision": true,
        "maxInputTokens": 128000,
        "maxOutputTokens": 16000
      }
    ]
  }
]

The apiKey value is the secret reference VS Code creates when you paste your Gateway Key. Keep the reference generated on your machine rather than copying the one above.

VS Code extensions (Continue and others)

For third-party OpenAI-compatible extensions (like Continue), set the provider to OpenAI, paste your Gateway Key, and set the API Base URL to the Gateway URL. In Continue's config file that looks like:

models:
  - name: Spineforge GPT-4o
    provider: openai
    model: openai/gpt-4o
    apiBase: https://spineforge-backend.onrender.com/gateway/v1
    apiKey: sfgw-xxxx-xxxx

Field names can vary between extension versions: if in doubt, use the extension's "OpenAI-compatible" settings.

Any other IDE or editor

The universal logic is simple:

  1. Select OpenAI (or OpenAI Compatible) as the provider.
  2. Paste the Gateway URL as the Base URL (sometimes called API Base, Endpoint, or Override URL).
  3. Use your Gateway Key as the API key.
  4. Define your model with a provider prefix, such as openai/gpt-4o.

This works for JetBrains-based and other editors whenever they expose a custom OpenAI-compatible endpoint. If a tool only accepts a fixed vendor endpoint with no override, it can't be routed through the gateway.

Antigravity IDE

Antigravity IDE is not supported yet. Use the Antigravity CLI integration instead.

Verify it works

  1. Send any chat message from the IDE.
  2. Open the Spineforge dashboard: the request should appear against your gateway key with its model, tokens, latency and cost.

Troubleshooting

SymptomLikely cause
401 / invalid API keyThe key isn't a valid gateway key (it must start with sfgw-), or it was typed with extra whitespace.
403 / forbiddenThe key was revoked, or its scopes don't allow that provider or model.
Model not foundThe model name is missing its provider prefix: e.g. use openai/gpt-4o, not gpt-4o, for names that contain a slash.
Provider key errorNo matching provider key in the Vault. Add it in the dashboard Vault.
Request never reaches the gatewayThe IDE still points at the vendor endpoint. Re-check the Base URL override.

Next steps