Skip to content

API-REFERENCE.md

Latest commit

 

History

History
115 lines (92 loc) · 2.96 KB

API-REFERENCE.md

File metadata and controls

115 lines (92 loc) · 2.96 KB
Raw

Proxy Endpoints

All endpoints at http://127.0.0.1:17077.

OpenAI Compatible

Base URL: http://127.0.0.1:17077/v1

Path Method Description
/v1/chat/completions POST Chat completions (stream + non-stream)
/v1/models GET List 44 available models
/v1/health GET Health check
/v1/token GET Get current access token
/v1/info GET Bridge info and config

Anthropic Compatible

Base URL: http://127.0.0.1:17077

Path Method Description
/v1/messages POST Messages API (stream + non-stream)
/messages POST Messages API (alternative path)

API Key

Any value works. The bridge uses your saved Command Code auth from ~/.commandcode/auth.json.

Client Configuration

OpenCode (OpenAI Compatible)

"Command Code": {
  "name": "Command Code",
  "options": {
    "baseURL": "http://127.0.0.1:17077/v1",
    "apiKey": "anything"
  },
  "models": {
    "deepseek/deepseek-v4-pro": {
      "name": "Command Code - DeepSeek V4 Pro /Op",
      "tool_call": true,
      "reasoning": true,
      "limit": { "context": 1048576, "input": 1048576, "output": 8192 }
    }
  }
}

OpenCode (Anthropic Compatible)

"Command Code Anthropic": {
  "npm": "@ai-sdk/anthropic",
  "name": "Command Code Anthropic",
  "options": {
    "baseURL": "http://127.0.0.1:17077/v1",
    "apiKey": "anything"
  },
  "models": {
    "deepseek/deepseek-v4-pro": {
      "name": "Command Code - DeepSeek V4 Pro /An",
      "tool_call": true,
      "reasoning": true,
      "limit": { "context": 1048576, "input": 1048576, "output": 8192 }
    }
  }
}

Zed (Anthropic Compatible)

Zed reads providers from settings.json under language_models.anthropic_compatible. The API key must be entered through Agent Settings (agent: open settings -> LLM Providers).

api_url must be the base URL without /v1 — Zed auto-appends /v1/messages.

"Anthropic Command Code (Go Plan) - Khip01 Local": {
  "api_url": "http://127.0.0.1:17077",
  "available_models": [
    {
      "name": "deepseek/deepseek-v4-pro",
      "display_name": "DeepSeek V4 Pro - CommandCode",
      "max_tokens": 1000000,
      "max_output_tokens": 64000,
      "extra_beta_headers": [],
      "capabilities": {
        "tools": true,
        "images": false,
        "prompt_caching": false
      }
    }
  ]
}

After saving, open Agent Settings, find the provider, enter any API key (the bridge uses your saved Command Code auth, the value does not matter), and the model will appear in the model dropdown.

Cost Tracking

To make your CLI agent report real Command Code costs, run:

commandcode-bridge cost-sync

or press [c] on the Cost Sync page in the TUI. This writes per-model cost fields (input, output, cache_read per 1M tokens) into your agent configs. The pricing table is validated against the live bridge /v1/models endpoint before syncing.