get_news_list

Instructions

Implementation Reference

• src/jnews_mcp_server/server.py:82-110 (handler)

  The actual async function that executes the 'get_news_list' tool logic. It calls the Juhe Toutiao API with type, page, page_size, and API key parameters, and returns news data as TextContent.

  async def get_news_list(type: str = "top", page: int = 1, page_size: int = 20) -> list[types.TextContent | types.ImageContent | types.EmbeddedResource]:
      """
      根据新闻类型获取今日热点新闻头条.
      """
      url = f"{JUHE_NEWS_API_BASE}/index"
      params = {
          "type": type,
          "page": page,
          "page_size": page_size,
          "key": JUHE_NEWS_API_KEY
      }
      async with httpx.AsyncClient() as client:
          response = await client.get(url, params=params)
          data = response.json()
          if data["error_code"] == 0:
              news_list = data["result"]["data"]
              return [
                  types.TextContent(
                      type="text",
                      text=f"{news_list}"
                  )
              ]
          else:
              return [
                  types.TextContent(
                      type="text",
                      text=f"Error: {data['reason']}"
                  )
              ]

• src/jnews_mcp_server/server.py:24-43 (registration)

  Tool registration using @server.list_tools() decorator. Defines the tool name 'get_news_list', its description, and JSON Schema input schema with 'type', 'page', and 'page_size' properties.

  return [
      types.Tool(
          name="get_news_list",
          description="通过新闻类型获取今日热点新闻头条",
          inputSchema={
              "type": "object",
              "properties": {
                  "type": {
                      "type": "string", "description": "新闻类型:top(推荐,默认),guonei(国内),guoji(国际),yule(娱乐),tiyu(体育),junshi(军事),keji(科技),caijing(财经),youxi(游戏),qiche(汽车),jiankang(健康)",
                  },
                  "page": {
                      "type": "number", "description": "当前页数, 默认1, 最大50"
                  },
                  "page_size": {
                      "type": "number", "description": "每页返回新闻条数, 默认20, 最大50"
                  },
              },
              # "required": ["type"],
          },
      ),

• src/jnews_mcp_server/server.py:28-42 (schema)

  Input schema for the 'get_news_list' tool. Defines 'type' (string), 'page' (number), and 'page_size' (number) as optional parameters with descriptions.

  inputSchema={
      "type": "object",
      "properties": {
          "type": {
              "type": "string", "description": "新闻类型:top(推荐,默认),guonei(国内),guoji(国际),yule(娱乐),tiyu(体育),junshi(军事),keji(科技),caijing(财经),youxi(游戏),qiche(汽车),jiankang(健康)",
          },
          "page": {
              "type": "number", "description": "当前页数, 默认1, 最大50"
          },
          "page_size": {
              "type": "number", "description": "每页返回新闻条数, 默认20, 最大50"
          },
      },
      # "required": ["type"],
  },

• src/jnews_mcp_server/server.py:65-67 (helper)

  Call tool dispatcher that routes 'get_news_list' requests to the handler function, extracting 'type' from arguments (defaulting to 'top').

  if name == "get_news_list":
      type_value = type_value = arguments.get("type", "top") if arguments else "top"
      return await get_news_list(type_value)