How to Use the Exa MCP in Google ADK

Q: How do I register the Exa tools in my Google ADK agent?

Instantiate McpToolset with your hosted MCP Server parameters and pass it to the LlmAgent constructor. The ADK handles the handshake and exposes the tools directly to Gemini.

Q: Can I restrict which Exa tools my Google ADK agent can access?

Yes, you can pass an optional toolnames filter to the ADK's toolset configuration. This lets you limit the agent to specific tools like searchkeyword or getcontents if you want to control API costs.

Q: How does Exa handle semantic search queries inside Google ADK?

The agent uses searchneural to execute semantic queries that understand user intent rather than just matching text strings. This is perfect for complex research tasks where the user's query is conversational.

Q: Can a Google ADK agent retrieve direct answers using Exa?

Yes, the agent can call the answer tool to get a synthesized response directly from the web search engine. This is ideal for quick lookups where you don't need to parse full page bodies.

Feed clean, real-time web context into Google ADK agents to ground 1M+ token Gemini reasoning with Exa search.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Exa MCP to Google ADK

Create your Vinkius account to connect Exa to Google ADK and route execution through our secure gateway. The platform manages server hosting, runtime updates, and security layers. Configuration requires no manual server provisioning.

GDPR Free for Subscribers

Setup Exa with Google ADK

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Grounding Gemini Context with Exa

The `search_neural` tool grounds Gemini's massive context window with real-time semantic web search results. By connecting the Exa MCP Server to your Google ADK setup, your agents can pull live web data directly into their context windows. This keeps your long-context reasoning grounded in real-world facts. Your agent can run queries using conceptual terms, bypassing the limitations of simple keyword matching. The ADK's native tools array takes this server's output and feeds it directly into your prompt context without manual parsing.

Targeted Domain Audits for Google ADK Enterprise Agents

The `search_domain` tool restricts your agent's research queries to a specific list of verified websites. This MCP Server allows enterprise search to look up internal documentation or trusted industry sources rather than the open web. When your agent finds a valuable resource, it can run `get_contents` to pull down clean, parsed text. This avoids wasting Gemini's 1M+ token context window on raw HTML, Javascript, or tracking scripts, saving you significant API costs.

Deep Similarity Audits on the Fly

The `find_similar` tool discovers related web properties based on a seed URL for instant market analysis. Your agent can analyze a competitor's site and instantly discover their entire market footprint without requiring human intervention. To make this even faster, the agent can use `find_similar_with_contents` to retrieve both the URLs and their parsed text in a single execution. This raw data can be piped directly into Vertex AI pipelines or BigQuery tables for further analysis.

Setup guide

Set up Exa MCP in Google ADK

Prerequisites

Python 3.10+ installed
google-adk package (pip install google-adk)
Active Vinkius subscription with a valid endpoint token

1

Install Google ADK

Run pip install google-adk to install the Agent Development Kit. MCP support is included via the McpToolset class.
2

Connect via SSE transport

Use McpToolset.from_server() with SseServerParams pointing to your Vinkius endpoint. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com.
3

Create an LlmAgent

Pass the returned mcp_tools list directly to LlmAgent(tools=mcp_tools). The ADK maps each MCP tool to a native Gemini function call — no manual schema definitions required.
4

Run with any Gemini model

The agent works with any Gemini model (gemini-2.0-flash, gemini-2.5-pro, etc.). Copy the full example on the right to get started with Exa tools in your ADK agent.

agent.py

from google.adk.agents import LlmAgent
from google.adk.tools.mcp_tool.mcp_toolset import McpToolset
from google.adk.tools.mcp_tool.mcp_session_manager import SseServerParams

# Connect to the MCP via SSE
mcp_tools, exit_stack = await McpToolset.from_server(
    connection_params=SseServerParams(
        url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
    )
)

# Create your agent with auto-discovered tools
agent = LlmAgent(
    name="Exa_agent",
    model="gemini-2.0-flash",
    instruction="You have access to Exa tools via MCP.",
    tools=mcp_tools,
)

Get your connection token →

Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Exa. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.

Why Choose Vinkius

Vinkius connects your tools to AI with real-time monitoring and automatic cost savings — all from one dashboard.

Connect Exa now

Real-time monitoring

Live

visibility into every interaction

Connect your favorite tools to your AI and see exactly what's happening — every request, every response, in real time.

Built-in savings

60%

lower AI costs

Vinkius compresses data between your apps and your AI automatically. Lower bills every month — no configuration required.

Single dashboard

One

place for every integration

Every tool your AI connects to, managed from a single screen. One account, complete control.

Common questions about Exa MCP in Google ADK

Instantiate `McpToolset` with your hosted MCP Server parameters and pass it to the `LlmAgent` constructor. The ADK handles the handshake and exposes the tools directly to Gemini.

Yes, you can pass an optional `tool_names` filter to the ADK's toolset configuration. This lets you limit the agent to specific tools like `search_keyword` or `get_contents` if you want to control API costs.

The agent uses `search_neural` to execute semantic queries that understand user intent rather than just matching text strings. This is perfect for complex research tasks where the user's query is conversational.

Yes, the agent can call the `answer` tool to get a synthesized response directly from the web search engine. This is ideal for quick lookups where you don't need to parse full page bodies.

All search terms and page URLs are transmitted over encrypted TLS connections directly to the Vinkius gateway. The server runs in a secure, isolated runtime where data is processed in-memory and immediately destroyed upon request completion.

Use it with your favorite AI tools

Connect this server to Cursor, Claude, VS Code, and more.

OpenAI Agents SDK sdk-python

Google ADK sdk-python

Pydantic AI sdk-python

Vercel AI SDK sdk-typescript