I built a LlamaIndex agent with real-time search using TalorData — here‘s how

July 16, 2026

The Problem

I‘ve been building AI agents with LlamaIndex, and I kept hitting the same wall: LLMs can’t access current information.

For a research assistant I was building, this was a dealbreaker. Users wanted answers about today‘s news, not 2023 data.

I looked at building my own scraping infrastructure, but that’s a rabbit hole I didn‘t want to go down. Maintaining proxies, parsing HTML, handling CAPTCHAs — life’s too short.

The Solution

I found TalorData SERP API, which provides structured search results from Google, Bing, Yandex, and DuckDuckGo through a single API. They announced official LlamaIndex integration in June 2026, so integration took about 20 minutes.

How It Works

1. Install packages:

bash

pip install llama-index-core llama-index-llms-openai talordata-serp

2. Create the search tool:

python

import os, json from llama_index.core.tools import FunctionTool from talordata_serp import TalorClient client = TalorClient(api_key=os.environ["TALORDATA_API_KEY"]) def search_web(query: str, engine: str = "google", num: int = 5) -> str: response = client.search(q=query, engine=engine, num=num, json=2) results = [ {"title": r.get("title"), "link": r.get("link"), "snippet": r.get("snippet")} for r in response.get("organic_results", [])[:num] ] return json.dumps(results, indent=2) search_tool = FunctionTool.from_defaults( fn=search_web, name="web_search", description="Search the web for real-time information." )

3. Build and run the agent:

python

from llama_index.core.agent import ReActAgent from llama_index.llms.openai import OpenAI llm = OpenAI(model="gpt-4o-mini", temperature=0) agent = ReActAgent.from_tools(tools=[search_tool], llm=llm, verbose=True) response = agent.chat("What's new in AI search engines in 2026?") print(response)