Skip to main content

VS Code with Continue and Keeptrusts

Continue supports an apiBase override for OpenAI-compatible models. Use its current config.yaml format; the older config.json format is deprecated.

Store the client token​

Continue's IDE extension resolves local secrets from a workspace .env, .continue/.env, or ~/.continue/.env. Put the client token in one of those git-ignored files:

KEEPTRUSTS_IDE_TOKEN=<client-api-token>

Restart VS Code after changing the secret.

Configure the model​

Add a model to Continue's config.yaml:

name: Keeptrusts
version: 1.0.0
schema: v1

models:
  - name: Keeptrusts Chat
    provider: openai
    model: gpt-5.4-mini
    apiBase: http://127.0.0.1:41002/v1
    apiKey: ${{ secrets.KEEPTRUSTS_IDE_TOKEN }}
    roles:
      - chat
      - edit
      - apply

Choose a model returned by the gateway's /v1/models endpoint. Do not assign the autocomplete role unless the upstream model and Continue configuration are appropriate for fill-in-the-middle completion.

See Continue's config reference for the current model fields and Continue's secret guidance for resolution order.

Verify​

Run kt events tail --since 10m --follow, select Keeptrusts Chat in Continue, and send a request. If Continue still calls another provider, check the active configuration and selected model rather than VS Code's generic proxy setting.

Next steps​