Here’s the table of contents:
MCP 协议 SSE 模式下的 HTTP 请求过程
下面演示了从连接MCP服务到执行资源获取、提示词获取、工具调用的完整HTTP交互过程。【备注:执行结果需要订阅1
中的数据流获取】
1. 建立 SSE 连接(GET)
curl -N -H "Accept: text/event-stream" "http://localhost:8000/mcp-chat-insight/sse"
2. 初始化(POST)
curl -X POST "http://localhost:8000/mcp-chat-insight/messages/?session_id=24f9f470dd0a4432a0acd4d5222d4d8a" \
-H "Content-Type: application/json" \
-d '{"method":"initialize","params":{"protocolVersion":"2025-03-26","capabilities":{},"clientInfo":{"name":"Postman Client - df1716d8-d6a3-4b88-8669-021162b72341","version":"1.0.0"}},"jsonrpc":"2.0","id":0}'
3. 通知已初始化(POST)
curl -X POST "http://localhost:8000/mcp-chat-insight/messages/?session_id=24f9f470dd0a4432a0acd4d5222d4d8a" \
-H "Content-Type: application/json" \
-d '{"method":"notifications/initialized","jsonrpc":"2.0"}'
4. 获取工具列表(POST)
curl -X POST "http://localhost:8000/mcp-chat-insight/messages/?session_id=24f9f470dd0a4432a0acd4d5222d4d8a" \
-H "Content-Type: application/json" \
-d '{"method":"tools/list","jsonrpc":"2.0","id":1}'
5. 获取资源列表(POST)
curl -X POST "http://localhost:8000/mcp-chat-insight/messages/?session_id=24f9f470dd0a4432a0acd4d5222d4d8a" \
-H "Content-Type: application/json" \
-d '{"method":"resources/list","jsonrpc":"2.0","id":2}'
6. 获取提示词列表(POST)
curl -X POST "http://localhost:8000/mcp-chat-insight/messages/?session_id=24f9f470dd0a4432a0acd4d5222d4d8a" \
-H "Content-Type: application/json" \
-d '{"method":"prompts/list","jsonrpc":"2.0","id":3}'
7. 获取资源模板列表(POST)
curl -X POST "http://localhost:8000/mcp-chat-insight/messages/?session_id=24f9f470dd0a4432a0acd4d5222d4d8a" \
-H "Content-Type: application/json" \
-d '{"method":"resources/templates/list","jsonrpc":"2.0","id":4}'
8. 调用 list_tables 工具(POST)
curl -X POST "http://localhost:8000/mcp-chat-insight/messages/?session_id=24f9f470dd0a4432a0acd4d5222d4d8a" \
-H "Content-Type: application/json" \
-d '{"method":"tools/call","params":{"name":"list_tables","arguments":{}},"jsonrpc":"2.0","id":5}'
PREVIOUSDBeaver中使用Nginx将Openai地址代理到自定义地址