opencode-commandcode is no longer needed — or recommended.

Command Code now offers an official API (the Provider API) that works natively with OpenCode. Every paid plan except the $1 Go plan includes API access, and connecting it takes about a minute — no plugin, no proxy, no browser-login tricks.

• The official API replaced us. GOAT ($10), Pro, Max, Team, and Provider plans can all call https://api.commandcode.ai/provider/v1 directly with a standard API key. OpenCode supports OpenAI-compatible providers out of the box, so a plugin adds nothing.
• The only audience left was Go-plan ($1) users bypassing a deliberate paywall. Command Code intentionally excludes the Go plan from API access (their API returns 403 upgrade_required). Using this plugin on a Go account violates the Command Code Terms of Service and accounts have been banned for it.
• Legal risk. Similar community projects (e.g. 9router) received DMCA takedowns for proxying Command Code's internal endpoints.

We don't want to put users' accounts — or this project — at risk. So the repository is archived.

1. Subscribe to any plan except Go (GOAT $10/mo and up), or the pay-as-you-go Provider plan ($15/mo).
2. Create an API key at Studio → API keys.
3. Add it to OpenCode — run opencode auth login, or add a custom provider to ~/.config/opencode/opencode.json:

{
  "$schema": "https://opencode.ai/config.json",
  "provider": {
    "commandcode": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "Command Code",
      "options": {
        "baseURL": "https://api.commandcode.ai/provider/v1",
        "apiKey": "{env:CMD_API_KEY}"
      },
      "models": {
        "zai-org/GLM-5.2": {},
        "deepseek/deepseek-v4-flash": {},
        "moonshotai/Kimi-K2.7-code": {}
      }
    }
  }
}

Then set CMD_API_KEY in your environment and run:

opencode run "Hello" --model commandcode/zai-org/GLM-5.2

The full live model catalog is at GET /provider/v1/models. Both OpenAI (/chat/completions) and Anthropic (/messages) endpoints are available, streaming included. See the Provider API docs for details, and Pricing & Limits for per-model rates.

This plugin provided browser OAuth and a local OpenAI-compatible proxy to Command Code's internal POST /alpha/generate endpoint, bringing Laguna and the live model catalog into OpenCode before an official API path existed. Its source remains available in this archived repository for reference, but it is unmaintained and will not receive updates.

Thanks to everyone who contributed, filed issues, and used it responsibly.

MIT