How to Use the ClickHouse (Vector Search) MCP in Pydantic AI

Q: How do I register the ClickHouse (Vector Search) MCP Server in Pydantic AI?

Use the unified MCPToolset class pointing to your Vinkius HTTP endpoint. Pass the toolset instance directly into your Agent constructor's toolsets parameter.

Q: Can I run raw SQL queries safely?

Yes. The executesql tool lets you run any query, but you should define a Pydantic model on your agent to validate the returned rows before using them.

Run type-safe ClickHouse (Vector Search) operations with Pydantic AI using this managed MCP Server for runtime validation.

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 Pydantic AI

Create your Vinkius account to connect ClickHouse (Vector Search) to Pydantic AI 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 Pydantic AI

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Type-safe vector queries with Pydantic AI

The `vector_search` tool runs distance-based similarity calculations and returns strictly typed search results. Because Pydantic AI validates the response at runtime, your agent never ingests malformed embedding vectors. If you need to run complex analytical queries, `execute_sql` lets the agent execute arbitrary DDL or SELECT statements. The output is parsed directly into your defined schemas, stopping raw database errors from crashing your app.

Structural schema extraction

The `describe_table` tool extracts column schemas and data types directly from your active database. This allows your agent to match its internal models against the real database structure before writing a query. To find where your vector data lives, the agent uses `list_databases` and `list_tables` to map out the active schemas. This prevents validation errors by making sure the agent only queries tables that actually exist.

Cluster state and index validation

The `get_table_stats` tool retrieves internal stats to check cluster health and table sizes. This gives your agent the metrics it needs to optimize query batch sizes dynamically. Using `get_version`, the agent checks for HNSW index support and engine capabilities. This ensures your code doesn't attempt vector operations on unsupported database versions. This MCP connection handles the version check automatically.

Setup guide

Set up ClickHouse (Vector Search) MCP in Pydantic AI

Prerequisites

Python 3.10+ installed
pydantic-ai-slim[fastmcp] package
Active Vinkius subscription with a valid endpoint token

1

Install Pydantic AI with FastMCP
Run pip install "pydantic-ai-slim[fastmcp]". The FastMCP toolset replaces the deprecated MCPServerHTTP class with full protocol support.
2

Configure the FastMCPToolset
Pass a JSON-style config dict to FastMCPToolset with your Vinkius URL. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com. Supports Streamable HTTP, SSE, and Stdio transports.
3

Create and run your agent
Pass the toolset to Agent(toolsets=[toolset]) and call agent.run(). Swap openai:gpt-4o for any supported model — Anthropic, Google, Mistral, or Groq.

agent.py

from pydantic_ai import Agent
from pydantic_ai.toolsets.fastmcp import FastMCPToolset

toolset = FastMCPToolset({
    "mcpServers": {
        "clickhouse-vector-search-mcp": {
            "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
        }
    }
})

agent = Agent(
    "openai:gpt-4o",
    toolsets=[toolset],
    system_prompt="You have access to ClickHouse (Vector Search) tools.",
)

result = await agent.run("List recent ClickHouse (Vector Search) transactions")
print(result.output)

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 Pydantic AI

Use the unified `MCPToolset` class pointing to your Vinkius HTTP endpoint. Pass the toolset instance directly into your Agent constructor's toolsets parameter.

The framework will raise a validation error immediately. This prevents silent failures or corrupted state, allowing your application to handle the error gracefully.

Yes. The server supports both Streamable HTTP and SSE transports. You can configure your connection method when instantiating the toolset.

Yes. The `execute_sql` tool lets you run any query, but you should define a Pydantic model on your agent to validate the returned rows before using them.

Vinkius processes all database calls in an ephemeral, zero-trust sandbox. Your connection details are completely hidden from the agent, preventing credential leaks or unauthorized access.

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