How does authentication work?
On first connection to an auth‑requiring server, you are prompted to open a URL in your browser to authorize. The program waits for the OAuth flow, then stores tokens in `~/.mcp-auth` for reuse.
@naotama2002
基本情報
This Go implementation allows MCP clients that only support local (stdio) connections to connect to remote MCP servers with authentication support.
以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。
{
"mcpServers": {
"mcp-remote-go": {
"command": "docker",
"args": [
"pull",
"ghcr.io/naotama2002/mcp-remote-go:latest"
]
}
}
}ツール一覧はまだありません
この掲載には表示できるツール一覧がまだありません。サーバーにツールがないという意味ではありません。対応機能はプロジェクトのドキュメントをご確認ください。
ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。
It is a Go implementation that proxies between a local MCP client using stdio transport and a remote MCP server using Server-Sent Events (SSE) with OAuth authentication. It enables MCP clients that only support local connections to reach remote servers securely.
Install by building from source or pulling the Docker image from GitHub Container Registry. Run as mcp-remote-go <remote-sse-url> [port] [--header ...] [--allow-http] or via Docker with docker run .... Configure MCP clients (Claude Desktop, Cursor, Windsurf) by editing their JSON config files to point to the binary or Docker command.
~/.mcp-auth for reuse.On first connection to an auth‑requiring server, you are prompted to open a URL in your browser to authorize. The program waits for the OAuth flow, then stores tokens in `~/.mcp-auth` for reuse.
Run `rm -rf ~/.mcp-auth` to delete all stored OAuth tokens.
Set the environment variable `SSL_CERT_FILE` to the path of your CA certificates file before running the program.
Pass the port number as a second argument; e.g., `mcp-remote-go https://remote.mcp.server/sse 9090`. In Docker, map the container port accordingly.
Yes, use the `--allow-http` flag. This should only be done on trusted internal networks, as HTTPS is normally required.

Handl speaks the Model Context Protocol. Your whole billing operation — projects, invoices, clients, reports, approvals — as tools any AI assistant can call, inside your autonomy rules.

B2B competitive intelligence, live sales battlecards, 5-axis ICP prospect fit scoring, and tech stack teardowns over MCP and REST with cited public-web evidence.

DeepMark helps teachers deliver rapid, consistent marking with meaningful feedback for every student — in a fraction of the time. What once took a week, now takes one free period.
🚀 The fast, Pythonic way to build MCP servers and clients.
Hindsight: Agent Memory That Learns
コメント