CommandCode Proxy Server (Node.js)

OpenAI-compatible proxy server for the CommandCode API, built in high-performance native Node.js (zero external dependencies). It exposes /v1/chat/completions, /v1/responses, and /v1/models endpoints so OpenAI-compatible clients (e.g. Claude Code, Aider, Continue, Roo Code) can call CommandCode models through a local HTTP server.

Repository: https://github.com/Rbil1y/command-code-proxy-server

Version: v1.0.8

Features

• OpenAI-compatible chat completions endpoint (/v1/chat/completions, /chat/completions)
• OpenAI Responses API endpoint (/v1/responses)
• Streaming (SSE) and non-streaming responses
• OpenAI-compatible model list endpoint (/v1/models)
• Health check endpoint (/health)
• Short model name mapping & automatic pass-through
• Automatic capping of max_tokens (200,000 max) to prevent upstream validation errors
• Optional default API key injection via CLI (-api-key)
• Per-request API key via Authorization header
• Configurable host and port
• Zero external node_modules dependencies (100% native Node.js)
• Integrated version check against GitHub tags and npm registry

Requirements

• Node.js 18.0.0 or newer

Run

npm start

or directly:

node bin/cli.js

Default server address:

http://127.0.0.1:55990

CLI options

npm start -- [options]
# or
node bin/cli.js [options]

Option	Default	Description
-host	127.0.0.1	Host to bind the server to
-port	55990	Port to run the server on
-api-key	empty	Optional default CommandCode API key to inject
-version	false	Print version and exit
-debug	false	Enable verbose debug logging
-help	false	Show help message

Examples:

# Run on default host and port
node bin/cli.js

# Inject your CommandCode API key automatically for all requests
node bin/cli.js -api-key your-commandcode-api-key

# Run on a custom port
node bin/cli.js -port 8080

# Expose on all network interfaces
node bin/cli.js -host 0.0.0.0

# Print version
node bin/cli.js -version

API key behavior

The proxy uses the API key in this order:

1. -api-key CLI value (if specified when starting the proxy server)
2. Authorization header from the incoming client request
3. If neither exists, the request returns 401 Unauthorized

Header format:

Authorization: Bearer your-commandcode-api-key

Supported model aliases

The proxy accepts full model IDs and short aliases:

Alias	Maps to
muse-spark-1.2-contributor, muse-contributor	meta/muse-spark-1.2-contributor
muse-spark-1.2, muse1.2	meta/muse-spark-1.2
muse-spark-1.1, muse-spark	meta/muse-spark-1.1
gpt-5.6-luna, luna	openai/gpt-5.6-luna
gpt-5.6-terra, terra	openai/gpt-5.6-terra
gpt-5.6-sol, sol	openai/gpt-5.6-sol
deepseek-v4-pro, deepseek-v4, deepseek-pro	deepseek/deepseek-v4-pro
deepseek-v4-flash, deepseek-flash	deepseek/deepseek-v4-flash
minimax-m2.7, minimax2.7	MiniMaxAI/MiniMax-M2.7
minimax-m2.5, minimax2.5, minimax	MiniMaxAI/MiniMax-M2.5
minimax-m3, minimax3	MiniMaxAI/MiniMax-M3
glm-5.2, glm5.2, glm52	zai-org/GLM-5.2
glm-5.1	zai-org/GLM-5.1
glm-5	zai-org/GLM-5
kimi-k3, kimi3	moonshotai/Kimi-K3
kimi-k2.6, kimi2.6	moonshotai/Kimi-K2.6
kimi-k2.5, kimi2.5	moonshotai/Kimi-K2.5
qwen-3.6-max-preview, qwen3.6-max	Qwen/Qwen3.6-Max-Preview
qwen-3.6-plus, qwen3.6-plus, qwen3.6	Qwen/Qwen3.6-Plus
step-3.5-flash, step3.5	stepfun/Step-3.5-Flash
step-3.7-flash, step3.7	stepfun/Step-3.7-Flash
gemini-3.1-flash-lite, gemini-flash-lite	google/gemini-3.1-flash-lite
qwen-3.7-max-free, qwen3.7-max-free	Qwen/Qwen3.7-Max-Free
qwen-3.7-max, qwen3.7-max	Qwen/Qwen3.7-Max
mimo-v2.5-pro, mimo-pro	xiaomi/mimo-v2.5-pro
mimo-v2.5, mimo	xiaomi/mimo-v2.5

Unknown model names are passed through unchanged.

Project structure

.
├── README.md
├── package.json
├── bin
│   └── cli.js
├── src
│   ├── convert.js
│   ├── index.js
│   ├── models.js
│   ├── proxy.js
│   ├── server.js
│   └── version.js
└── test
    ├── convert.test.js
    └── models.test.js

License

MIT