Google Search

The Google Search tool connects your AI agent directly to Google's real-time web search index, grounding responses in up-to-the-minute facts and verified sources.

Tool Identifier: search-google

Why Use Search Grounding?

Large language models have a fixed knowledge cutoff date. When users ask about today's events, breaking news, recent product updates, or real-time sports scores, an ungrounded model will either apologize or hallucinate plausible-sounding guesses.

With Google Search Grounding:

  • Real-Time Freshness: Retrieve news, announcements, and events happening right now.
  • Fact Verification: Reduce hallucinations by anchoring generated text in authoritative web search results.
  • Source Citations: Responses include inline reference links so users can inspect and trust the primary sources.

How to Enable

In the UI

When configuring an agent in the Agent Builder or editing an existing agent, check the box for: ☑️ Google Search 🔎

In Agent YAML

Add search-google to the tools array:

tools:
  - search-google

Example Prompts & Use Cases

  • Market Intelligence: "What were the key announcements from Google I/O this year?"
  • Current Events: "What is happening in the world today?"
  • Competitive Analysis: "Compare the latest pricing plans of popular cloud hosting services."
  • Fact Checking: "When is the next solar eclipse visible from Central Europe?"

Try It in Practice

The default Google Gemini Agent has Google Search grounding pre-enabled.

  1. Open Chat with Google Gemini.
  2. Click the quick-start prompt: "What's new in the world today?".
  3. Notice how the agent searches the web, summarizes current headlines, and cites its sources!