Provides tools for searching the web using Brave Search engine.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@mcp-gateway search the web for MCP protocol"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
複数の子 MCP サーバーを束ねて、単一の MCP サーバーとして公開するゲートウェイ。
llama-server -m model.gguf --ui-mcp-proxyRelated MCP server: Agent Aggregator
llama-server（または任意の MCP クライアント）が POST /mcp に接続し、ツール一覧取得・実行を行う。ゲートウェイは子 MCP サーバー群にリクエストをルーティングし、結果を返す。
MCPクライアント（llama-server など）
↓ MCP over HTTP (POST /mcp)
mcp-gateway :8787
↙ ↓ ↘
filesystem brave ...子MCPサーバー群cp .env.example .env # 必要に応じて PORT を変更
mkdir -p data # filesystem-mcp の対象ディレクトリ
docker compose up --buildservers.json（プロジェクトルート。data/ の中ではない）にブロックを追加するだけ:
{
"filesystem": { "command": "npx", "args": ["-y", "@modelcontextprotocol/server-filesystem", "/data"] },
"brave": { "command": "npx", "args": ["-y", "@modelcontextprotocol/server-brave-search"], "env": { "BRAVE_API_KEY": "..." } }
}メソッド | パス | 説明 |
POST |
| MCP JSON-RPC（リクエスト・通知） |
# initialize
curl http://localhost:8787/mcp \
-H 'content-type: application/json' \
-d '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2024-11-05","capabilities":{},"clientInfo":{"name":"test","version":"0"}}}'
# tools/list
curl http://localhost:8787/mcp \
-H 'content-type: application/json' \
-d '{"jsonrpc":"2.0","id":2,"method":"tools/list"}'
# tools/call（ツール名は <サーバー名>__<ツール名> の形式）
curl http://localhost:8787/mcp \
-H 'content-type: application/json' \
-d '{"jsonrpc":"2.0","id":3,"method":"tools/call","params":{"name":"filesystem__read_file","arguments":{"path":"/data/hello.txt"}}}'pnpm install
pnpm test # vitest
pnpm lint # biome
pnpm dev # tsx で起動This server cannot be deployed
MCP server for OnceAsk, the AI-native current-address layer for people and agents.
Search, vet & assemble MCP servers from your agent: verified tools, risk labels, and trust scores.
Trust, freshness, policy, and discovery layer for public MCP servers.
