InternetSearch
Perform real-time internet searches to retrieve answers for specific queries, leveraging the Bocha Search API for accurate and relevant results.
Instructions
联网搜索对应问题的答案
Args: query: 需要联网搜索的问题
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| query | Yes | ||
| txt_count | No |
Implementation Reference
- Internet_search.py:16-18 (registration)FastMCP server initialization with name "Internet-search" and @mcp.tool() decorator which registers the InternetSearch tool.mcp = FastMCP("Internet-search") @mcp.tool()
- Internet_search.py:19-61 (handler)The handler function for the InternetSearch tool. It performs an HTTP POST request to the bochaai.com search API with the query, extracts summaries from the search results, and returns formatted text with reference IDs.def InternetSearch(query,txt_count=5): """联网搜索对应问题的答案 Args: query: 需要联网搜索的问题 """ headers = { "Authorization": f"Bearer {SEARCH_API_KEY}", # 替换为你的实际 API Key "Content-Type": "application/json" } payload = { "query": f"{query}", "freshness": "noLimit", "count": txt_count, "answer": False, "stream": False } Webtxt="" try: print("开始联网搜索") response = requests.post( "https://api.bochaai.com/v1/ai-search", headers=headers, json=payload ) response.raise_for_status() # 检查 HTTP 错误 i=0 for value in json.loads((response.json()["messages"][0]["content"]))["value"]: Webtxt=Webtxt+f"参考资料id：{i}\n"+value["summary"]+"\n" i+=1 #为了兼容web-search添加的代码 # for value in response.json()["data"]["webPages"]["value"]: # Webtxt=Webtxt+f"参考资料id：{i}\n"+value["snippet"]+"\n" # i+=1 # 解析并叠加响应 except requests.exceptions.RequestException as e: print(f"Request failed: {e}") except ValueError as e: print(f"Failed to parse JSON response: {e}") return Webtxt