初次提交

sci_mcp/general_mcp/searxng_query/searxng_query_tools.py

@@ -0,0 +1,78 @@
+"""
+Search Online Module
+
+This module provides functions for searching information on the web.
+"""
+from ...core.llm_tools import llm_tool
+from ...core.config import general_config
+import asyncio
+import os
+from typing import Annotated, Any, Dict, List, Union
+from langchain_community.utilities import SearxSearchWrapper
+import mcp.types as types
+
+
+@llm_tool(name="search_online_searxng", description="Search scientific information online using searxng")
+async def search_online_searxng(
+    query: Annotated[str, "Search term"],
+    num_results: Annotated[int, "Number of results"] = 5
+) -> str:
+    """
+    Searches for scientific information online and returns results as a formatted string.
+
+    Args:
+        query: Search term for scientific content
+        num_results: Number of results to return
+
+    Returns:
+        Formatted string with search results (titles, snippets, links)
+    """
+    # lzy: 此部分到正式发布时可能要删除，因为searxng 已在本地部署，因此本地调试时无需设置代理
+    os.environ['HTTP_PROXY'] = ''
+    os.environ['HTTPS_PROXY'] = ''
+    try:
+        max_results = min(int(num_results), general_config.SEARXNG_MAX_RESULTS)
+        search = SearxSearchWrapper(
+            searx_host=general_config.SEARXNG_HOST,
+            categories=["science",],
+            k=num_results
+        )
+
+        # Execute search in a separate thread to avoid blocking the event loop
+        # since SearxSearchWrapper doesn't have native async support
+        loop = asyncio.get_event_loop()
+        raw_results = await loop.run_in_executor(
+            None,
+            lambda: search.results(query, language=['en','zh'], num_results=max_results)
+        )
+
+        # Transform results into structured format
+        formatted_results = []
+        for result in raw_results:
+            formatted_results.append({
+                "title": result.get("title", ""),
+                "snippet": result.get("snippet", ""),
+                "link": result.get("link", ""),
+                "source": result.get("source", "")
+            })
+
+        # Format results into a readable Markdown string
+        result_str = f"Search Results for '{query}' ({len(formatted_results)} items):\n\n"
+        if len(formatted_results) > 0:
+            for i, res in enumerate(formatted_results):
+                title = res.get("title", "No Title")
+                snippet = res.get("snippet", "No Snippet")
+                link = res.get("link", "No Link")
+                source = res.get("source", "No Source")
+                result_str += f"{i + 1}. **{title}**\n"
+                result_str += f"   - Snippet: {snippet}\n"
+                result_str += f"   - Link: [{link}]({link})\n"
+                result_str += f"   - Source: {source}\n\n"
+        else:
+            result_str += "No results found.\n"
+
+        return result_str
+
+    except Exception as e:
+        return f"Error: {str(e)}"
+