OpenCode

Skip it — one-line setup

The script installs OpenCode from its official source, writes the gateway settings into the tool's own config file (backing up any existing one) and runs a live check. It asks for your gateway key, or take it from --key / --key-file. --uninstall restores exactly what it wrote.

curl -fsSL https://quantoport.com/agents/opencode/setup | bash

1 · Install

Install OpenCode from its own official source.

curl -fsSL https://opencode.ai/install | bash   # macOS/Linux; cross-platform: npm install -g opencode-ai

2 · Configure — point it at QuantoPort

Gateway base URL https://quantoport.com; get your key from the console.

1) In the TUI run /connect, scroll to Other, enter a provider id (e.g. quantoport) and your API key (stored in ~/.local/share/opencode/auth.json).
2) Add the provider in opencode.json (project root) or the global ~/.config/opencode/opencode.json:

{
  "$schema": "https://opencode.ai/config.json",
  "provider": {
    "quantoport": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "QuantoPort",
      "options": { "baseURL": "https://quantoport.com/v1" },
      "models": { "<model-id>": { "name": "<Model name>" } }
    }
  }
}

On the China endpoint set options.baseURL to https://api.shinlinkai.com/v1 (OpenAI wire only).
3) Run /models and pick the model.

3 · Verify

One request proves you are on our channel.

opencode run "reply with OK"      # or opencode auth list to inspect credentials

4 · Common pitfalls

• baseURL must include the /v1 version segment.
• Use npm @ai-sdk/openai-compatible (drives /v1/chat/completions); use @ai-sdk/openai only if the gateway speaks /v1/responses.
• The global ~/.config/opencode/opencode.json and the project opencode.json are merged; later sources override conflicting keys. OPENCODE_CONFIG can point at a custom config.
• opencode auth login is equivalent to /connect; credentials live in ~/.local/share/opencode/auth.json.