How to Use the ClickHouse (Vector Search) MCP in LlamaIndex

Q: Can I use ClickHouse (Vector Search) as an MCP tool in LlamaIndex agents?

Yes, use BasicMCPClient to connect to the server and wrap it in McpToolSpec. You can then pass the tools to your FunctionAgent for dynamic execution.

Q: How does LlamaIndex index the outputs of the ClickHouse (Vector Search) MCP Server?

The agent calls tools like vectorsearch or executesql and stores the returned datasets in memory. LlamaIndex can then index these results for follow-up semantic queries.

Q: Can I restrict which ClickHouse (Vector Search) MCP Server tools the LlamaIndex agent can access?

Yes, you can use the allowedtools filter when setting up your tool specification. This lets you restrict the agent to read-only tools like vectorsearch while blocking executesql.

Q: How does the agent handle complex table schemas?

The agent calls describetable to pull the active column schemas. It uses this structural data to format its SQL queries correctly before sending them to the database.

Feed high-speed ClickHouse vector search results directly into your LlamaIndex RAG pipelines.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect ClickHouse (Vector Search) MCP to LlamaIndex

Create your Vinkius account to connect ClickHouse (Vector Search) to LlamaIndex 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 ClickHouse (Vector Search) with LlamaIndex

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Ground LlamaIndex RAG with ClickHouse MCP Server data

The `vector_search` tool queries live vector embeddings to ground your RAG pipelines. By connecting this MCP Server to LlamaIndex, your agent queries live vector embeddings directly. The output instantly merges into your query pipeline, grounding responses in fresh database records instead of static files. The agent checks active cluster limits and configurations using `get_version` to adapt its search strategy. This ensures you are using the fastest available distance metrics on your index.

Index structural metadata dynamically

The `list_databases` tool maps out the structure of your ClickHouse instance dynamically. Your index needs to know the shape of your data, and this tool lets the agent discover tables without manual schema definitions. It can then pull specific records without guessing table names or columns. When it needs deeper structural insights, the agent calls `describe_table`. This extracts the exact active column schemas, allowing LlamaIndex to build accurate metadata filters for your search queries.

Run arbitrary SQL for advanced data retrieval

The `execute_sql` tool runs complex SELECT statements or DDL operations directly. This combines structured SQL queries with unstructured vector searches in the same pipeline. You get the speed of ClickHouse vectors with the precision of relational filters. To prevent performance bottlenecks, the agent monitors table sizes and cluster health using `get_table_stats`. If a table grows too large, the agent can adjust its batch sizes or alert your monitoring system.

Setup guide

Set up ClickHouse (Vector Search) MCP in LlamaIndex

Prerequisites

Python 3.10+ installed
llama-index-tools-mcp package
Active Vinkius subscription with a valid endpoint token

1

Install dependencies
Run pip install llama-index-tools-mcp llama-index-llms-openai. The MCP tools package provides BasicMCPClient and McpToolSpec.
2

Connect with BasicMCPClient
Point BasicMCPClient to your Vinkius endpoint URL. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com. Supports SSE and Streamable HTTP transports.
3

Convert to LlamaIndex tools
Call mcp_tool_spec.to_tool_list_async() to convert all ClickHouse (Vector Search) MCP tools into native FunctionTool objects that any LlamaIndex agent can use.
4

Run with any LLM
Create a FunctionAgent with the tools and your preferred LLM. Swap OpenAI for Anthropic, Gemini, or any LlamaIndex-supported provider.

agent.py

from llama_index.tools.mcp import BasicMCPClient, McpToolSpec
from llama_index.core.agent.workflow import FunctionAgent
from llama_index.llms.openai import OpenAI

# Connect to the MCP
mcp_client = BasicMCPClient(
    "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
)
mcp_tool_spec = McpToolSpec(client=mcp_client)

# Convert MCP tools to LlamaIndex tools
tools = await mcp_tool_spec.to_tool_list_async()

# Create and run the agent
agent = FunctionAgent(
    tools=tools,
    llm=OpenAI(model="gpt-4o"),
    system_prompt="You have access to ClickHouse (Vector Search) tools.",
)
response = await agent.run("List recent ClickHouse (Vector Search) data")

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 ClickHouse. 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 ClickHouse (Vector Search) 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 ClickHouse (Vector Search) MCP in LlamaIndex

Yes, use `BasicMCPClient` to connect to the server and wrap it in `McpToolSpec`. You can then pass the tools to your `FunctionAgent` for dynamic execution.

The agent calls tools like `vector_search` or `execute_sql` and stores the returned datasets in memory. LlamaIndex can then index these results for follow-up semantic queries.

Yes, you can use the `allowed_tools` filter when setting up your tool specification. This lets you restrict the agent to read-only tools like `vector_search` while blocking `execute_sql`.

The agent calls `describe_table` to pull the active column schemas. It uses this structural data to format its SQL queries correctly before sending them to the database.

Your credentials are encrypted and stored inside the secure Vinkius vault. The MCP Server runs in a zero-trust sandbox, so the agent never has direct access to your database password.

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