How to Use the Elasticsearch Vector MCP in Google ADK

Q: How can a Google ADK agent run a search against Elasticsearch Vector?

You equip your LlmAgent with this MCP toolset. The agent can then invoke the search tool, providing a vector embedding. The server queries Elasticsearch and returns the most similar documents, which your agent can then use in its reasoning process.

Q: What's the process for creating an index in Elasticsearch Vector using the Google ADK?

Your agent calls the createindex tool. You'll need to provide a name and a mapping that specifies the densevector type and its dimensions, matching the output of your embedding model, like one from Vertex AI.

Connect your Google ADK agent to Elasticsearch for vector search, turning your Gemini-powered insights into action.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Elasticsearch Vector MCP to Google ADK

Create your Vinkius account to connect Elasticsearch Vector 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 Elasticsearch Vector with Google ADK

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Integrate Vector Search into Google Cloud

This server gives your Google ADK agent direct control over vector search in Elasticsearch. You can take embeddings generated by Vertex AI, for example, and use the `index_document` tool to load them straight into your search index. It works the other way, too. An agent can use the `search` tool to find relevant documents in Elasticsearch, then use that data as context for a query to BigQuery or another Google service. It bridges your search cluster with the rest of your GCP stack.

Manage Indexes with Your Google ADK Agent

Forget manual index management. Your agent can programmatically run `create_index` to define a new `dense_vector` index, complete with the right dimensions for your models. This is essential for building automated data pipelines. Your agent can also maintain its environment. Use `list_indexes` to get a directory of available indexes and `get_index` to inspect a specific one's mapping. This lets your Gemini-powered agent reason about the state of your Elasticsearch cluster before it acts. This is a core function of this MCP Server.

Build Long-Context Enterprise Agents

The Google ADK is built for complex tasks using Gemini's long context window. Your agent can pull large amounts of data, run a `search` on Elasticsearch to find relevant vectors, and use the results to inform its next steps in a long-running job. You can build agents that, for instance, monitor a data stream, use `index_document` to update an Elasticsearch index in real-time, and `delete_document` to prune old entries. This MCP toolset gives your agent the basic verbs it needs to manage a knowledge base.

Setup guide

Set up Elasticsearch Vector 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 Elasticsearch Vector 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="Elasticsearch Vector_agent",
    model="gemini-2.0-flash",
    instruction="You have access to Elasticsearch Vector 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 Elasticsearch Vector. 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 Elasticsearch Vector 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 Elasticsearch Vector MCP in Google ADK

You equip your `LlmAgent` with this MCP toolset. The agent can then invoke the `search` tool, providing a vector embedding. The server queries Elasticsearch and returns the most similar documents, which your agent can then use in its reasoning process.

Your agent calls the `create_index` tool. You'll need to provide a name and a mapping that specifies the `dense_vector` type and its dimensions, matching the output of your embedding model, like one from Vertex AI.

Yes. The agent can query BigQuery for metadata, generate embeddings, and then use the `index_document` tool to push that structured data into Elasticsearch Vector. It acts as the bridge between your data warehouse and your search index.

Yes, the Google ADK supports this directly. When you configure the `McpToolset`, you can use the `tool_names` filter to provide an explicit list of tools, like `['search', 'get_index']`, that you want to expose to the agent.

The MCP server acts as a secure proxy. Your agent's requests, including document data and query vectors, are sent over an encrypted connection to our isolated environment and forwarded to your Elasticsearch cluster. We don't log or persist any of this payload data.

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