Tool setup

Continue

Continue, the open-source assistant for VS Code and JetBrains, treats AdvancedMind AI as an OpenAI-compatible provider: set the base URL, your key, and a model id.

Before you start

You need an active API key with credits on the org. Create one under API keys and copy the secret; it is shown once. A standard-scoped key can use either Klara preview whenever that route is available.

1. Install Continue

Install the Continue extension from the VS Code Marketplace, or from the JetBrains Marketplace inside your IDE. Open the Continue panel once so it generates a default config, which you will edit in the next step.

2. Configure the provider

Continue reads a per-assistant config. In recent versions this is a config.yaml at ~/.continue/config.yaml; older builds use config.json at ~/.continue/config.json. Either way, declare each model with provider: openai so Continue uses its OpenAI-compatible client, and point apiBase at the gateway.

config.yaml

Store the secret in Continue's secrets rather than inline, and reference it with ${{ secrets.AMAI_API_KEY }} so the key never sits in a checked-in file:

~/.continue/config.yaml

name: AdvancedMind AI
version: 1.0.0
schema: v1
models:
  - name: Klara Base Preview v1
    provider: openai
    model: Klara-base-preview-v1
    apiBase: https://api.advancedmind.ai/v1
    apiKey: ${{ secrets.AMAI_API_KEY }}
    roles:
      - chat
      - edit
      - apply
  - name: Klara Base Preview v2
    provider: openai
    model: Klara-base-preview-v2
    apiBase: https://api.advancedmind.ai/v1
    apiKey: ${{ secrets.AMAI_API_KEY }}
    roles:
      - chat
      - edit
      - apply

config.json

On the older JSON config, set the same three fields per model: apiBase, apiKey, and model. Paste the amai_sk_... secret in place of the placeholder:

~/.continue/config.json

{
  "models": [
    {
      "title": "Klara Base Preview v1",
      "provider": "openai",
      "model": "Klara-base-preview-v1",
      "apiBase": "https://api.advancedmind.ai/v1",
      "apiKey": "amai_sk_..."
    },
    {
      "title": "Klara Base Preview v2",
      "provider": "openai",
      "model": "Klara-base-preview-v2",
      "apiBase": "https://api.advancedmind.ai/v1",
      "apiKey": "amai_sk_..."
    }
  ]
}

The apiBase must end in /v1; Continue appends /chat/completions to it. Both keys above use the same secret, so a single entry in your secrets store covers every model. Do not commit a config that has the raw key inlined.

3. Choose a model

Klara Base Preview v2: use it first for harder repository analysis, debugging, tests, and implementation loops.
Klara Base Preview v1: a more direct default for chat and focused edits.

The wire ids for both Klara previews are in the config above. Assign chat, edit, and apply roles to either preview, then switch between v1 for directness and v2 for harder work. Full pricing is on the pricing & credits page.

Tool and agent use

Continue's Agent mode lets the model call tools: read and edit files, run terminal commands, search the workspace. The Klara previews support the OpenAI tool-calling format, so Agent mode works once a model is set. Give the agent the edit and apply roles so it can write changes back, and review each proposed diff before applying it. An agent loop sends your files and intermediate results back as context on every step, so input tokens and cost grow with the length of the run.

Cost visibility

Every request from Continue is metered against your org balance and recorded with its model, token counts, and final cost on the Usage page. Top up or set a low-credit alert on Billing. A request that would run past your balance is blocked before any provider call.

Key scopes and limits

Give Continue its own key so you can revoke it independently of other tools. A standard key is enough for ordinary Klara work; use a research key only for an approved Research Mode scope. Set a monthly spend limit when you create the key to cap an unattended agent run. Manage and revoke keys on the API keys page.

Next: streaming, error handling, or the OpenWebUI setup guide.