Add tools for searching arXiv, web, and Wikipedia

tools/arxiv_tool.py

@@ -0,0 +1,73 @@
+import importlib
+
+__all__ = ['ArxivTool']
+
+
+class ArxivTool():
+    dependencies = ["arxiv==2.1.3"]
+
+    inputSchema = {
+        "name": "ArxivTool",
+        "description": "Searches arXiv for academic papers based on a query.",
+        "parameters": {
+            "type": "object",
+            "properties": {
+                "query": {
+                    "type": "string",
+                    "description": "Search query for papers (e.g., 'superconductors gem5').",
+                },
+                "max_results": {
+                    "type": "integer",
+                    "description": "Maximum number of papers to retrieve. Default is 5.",
+                    "default": 5
+                }
+            },
+            "required": ["query"],
+        }
+    }
+
+    def __init__(self):
+        pass
+
+    def run(self, **kwargs):
+        query = kwargs.get("query")
+        max_results = kwargs.get("max_results", 5)
+
+        if not query:
+            return {
+                "status": "error",
+                "message": "Missing required parameter: 'query'",
+                "output": None
+            }
+
+        try:
+            arxiv = importlib.import_module("arxiv")
+            client = arxiv.Client()
+
+            search = arxiv.Search(
+                query=query,
+                max_results=max_results,
+            )
+
+            papers = []
+            for result in client.results(search):
+                papers.append({
+                    "title": result.title,
+                    "authors": [author.name for author in result.authors],
+                    "published": result.published.isoformat(),
+                    "summary": result.summary.strip(),
+                    "pdf_url": result.pdf_url,
+                })
+
+            return {
+                "status": "success",
+                "message": f"Found {len(papers)} paper(s) on arXiv",
+                "output": papers,
+            }
+
+        except Exception as e:
+            return {
+                "status": "error",
+                "message": f"arXiv search failed: {str(e)}",
+                "output": None,
+            }

tools/web_search_tool.py

@@ -0,0 +1,65 @@
+import importlib
+
+__all__ = ['WebSearchTool']
+
+
+class WebSearchTool():
+    dependencies = ["googlesearch-python==1.3.0"]
+
+    inputSchema = {
+        "name": "WebSearchTool",
+        "description": "Searches a specific website for a given query using Google search.",
+        "parameters": {
+            "type": "object",
+            "properties": {
+                "website": {
+                    "type": "string",
+                    "description": "The website domain to search in (e.g., 'stackoverflow.com').",
+                },
+                "query": {
+                    "type": "string",
+                    "description": "The query string to search for on the website.",
+                }
+            },
+            "required": ["website", "query"],
+        }
+    }
+
+    def __init__(self):
+        pass
+
+    def run(self, **kwargs):
+        print("Running web search")
+
+        website = kwargs.get("website")
+        query = kwargs.get("query")
+
+        if not website or not query:
+            return {
+                "status": "error",
+                "message": "Missing required parameters: 'website' and 'query'",
+                "output": None
+            }
+
+        search_query = f"site:{website} {query}"
+        results = []
+
+        googlesearch = importlib.import_module("googlesearch")
+
+        try:
+            for result in googlesearch.search(search_query, num_results=10):
+                if "/search?num=" not in result:
+                    results.append(result)
+
+            return {
+                "status": "success",
+                "message": "Search completed successfully",
+                "output": results,
+            }
+
+        except Exception as e:
+            return {
+                "status": "error",
+                "message": f"Search failed: {str(e)}",
+                "output": None,
+            }

tools/wikipedia_tool.py

@@ -0,0 +1,102 @@
+import importlib
+
+__all__ = ['WikipediaTool']
+
+
+class WikipediaTool():
+    dependencies = ["requests==2.32.3", "beautifulsoup4==4.13.3"]
+
+    inputSchema = {
+        "name": "WikipediaTool",
+        "description": "Searches Wikipedia for a given question and returns a short summary.",
+        "parameters": {
+            "type": "object",
+            "properties": {
+                "question": {
+                    "type": "string",
+                    "description": "The topic or question to search on Wikipedia.",
+                }
+            },
+            "required": ["question"],
+        }
+    }
+
+    def __init__(self):
+        pass
+
+    def run(self, **kwargs):
+        question = kwargs.get("question")
+        if not question:
+            return {
+                "status": "error",
+                "message": "Missing required parameter: 'question'",
+                "output": None
+            }
+
+        print(f"Searching Wikipedia for: {question}")
+
+        requests = importlib.import_module("requests")
+        bs4 = importlib.import_module("bs4")
+        BeautifulSoup = bs4.BeautifulSoup
+
+        search_url = "https://en.wikipedia.org/w/api.php"
+        search_params = {
+            "action": "query",
+            "format": "json",
+            "list": "search",
+            "srsearch": question,
+            "srlimit": 1,
+        }
+
+        try:
+            response = requests.get(search_url, params=search_params)
+            if response.status_code != 200:
+                return {
+                    "status": "error",
+                    "message": "Wikipedia API request failed.",
+                    "output": None,
+                }
+
+            data = response.json()
+            search_results = data.get("query", {}).get("search", [])
+
+            if not search_results:
+                return {
+                    "status": "error",
+                    "message": "No results found on Wikipedia.",
+                    "output": None,
+                }
+
+            top_result = search_results[0]["title"]
+            page_url = f"https://en.wikipedia.org/wiki/{top_result.replace(' ', '_')}"
+            print(f"Fetching full content from: {page_url}")
+
+            html_url = f"https://en.wikipedia.org/api/rest_v1/page/html/{top_result.replace(' ', '_')}"
+            html_response = requests.get(html_url)
+
+            if html_response.status_code != 200:
+                return {
+                    "status": "error",
+                    "message": "Failed to fetch article content.",
+                    "output": None,
+                }
+
+            soup = BeautifulSoup(html_response.text, "html.parser")
+            paragraphs = [p.get_text() for p in soup.find_all("p") if p.get_text()]
+            full_text = " ".join(paragraphs)
+            summary = " ".join(full_text.split(". ")[:5])  # First 5 sentences
+
+            output_text = f"**{top_result}**\n{summary}...\n[Read more]({page_url})"
+
+            return {
+                "status": "success",
+                "message": "Wikipedia article summary retrieved successfully.",
+                "output": output_text,
+            }
+
+        except Exception as e:
+            return {
+                "status": "error",
+                "message": f"Exception occurred: {str(e)}",
+                "output": None,
+            }