The best way to deploy mcp server. A high-performance WebSocket/SSE transport layer & gateway for Anthropic's MCP (Model Context Protocol) — powered by Nginx, Nchan, and FastAPI.
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add nchan-transport -- python -m httmcp -f gpt_actions_openapi.json -p http://nchan:80~/.codex/config.toml
[mcp_servers.nchan-transport]
command = "python"
args = ["-m", "httmcp", "-f", "gpt_actions_openapi.json", "-p", "http://nchan:80"]
# codex mcp add nchan-transport -- python -m httmcp -f gpt_actions_openapi.json -p http://nchan:80~/.cursor/mcp.json
{
"mcpServers": {
"nchan-transport": {
"command": "python",
"args": [
"-m",
"httmcp",
"-f",
"gpt_actions_openapi.json",
"-p",
"http://nchan:80"
]
}
}
}.vscode/mcp.json
{
"servers": {
"nchan-transport": {
"type": "stdio",
"command": "python",
"args": [
"-m",
"httmcp",
"-f",
"gpt_actions_openapi.json",
"-p",
"http://nchan:80"
]
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"nchan-transport": {
"command": "python",
"args": [
"-m",
"httmcp",
"-f",
"gpt_actions_openapi.json",
"-p",
"http://nchan:80"
]
}
}
}claude_desktop_config.json
{
"mcpServers": {
"nchan-transport": {
"command": "python",
"args": [
"-m",
"httmcp",
"-f",
"gpt_actions_openapi.json",
"-p",
"http://nchan:80"
]
}
}
}.trae/mcp.json
{
"mcpServers": {
"nchan-transport": {
"command": "python",
"args": [
"-m",
"httmcp",
"-f",
"gpt_actions_openapi.json",
"-p",
"http://nchan:80"
]
}
}
}