Sign in Book a demo
Back to Tools
Databases API Key
Build Your Agent View Clickhouse Docs
clickhouse logo

Clickhouse MCP Tools for AI Agents

Let an agent discover databases and tables, inspect table schemas, and run read-only ClickHouse queries.

Available Tools

5 tools available in this integration

Clickhouse.DiscoverDatabases

Discover all the databases in the ClickHouse database.

Clickhouse.DiscoverSchemas

Discover all the schemas in the ClickHouse database.

Clickhouse.DiscoverTables

Discover all the tables in the ClickHouse database when the list of tables is not known.

Clickhouse.ExecuteSelectQuery

You have a connection to a ClickHouse database.

Clickhouse.GetTableSchema

Get the schema/structure of a ClickHouse table in the ClickHouse database

Securing Clickhouse

The more capabilities agents get, the more useful they are. However, more capabilities bring more risks of undesired effects such as data loss and exfiltration.

This is why Arcade provides you with contextual rules that dictate how and when a tool is allowed to run. Arcade’s contextual access lets you govern tool visibility and behavior at every stage of execution and control who can see a tool, what inputs are allowed, and what comes back.

Learn about Contextual Access Policies

For Clickhouse, you can implement some of these policies:

Access when tools are listed for a user
  • Decide which tools an agent is allowed to see at all. A tool that is never listed cannot be called, whoever is asking.
Pre-execution before each tool call
  • Inspect the inputs a call is about to run with, and block it or send it for human approval when they fall outside what the agent should do unsupervised.
Post-execution after the tool returns
  • Redact email addresses, phone numbers, access tokens, and other customer identifiers from Clickhouse.ExecuteSelectQuery results before they reach the model.
  • Strip column comments and default expressions from Clickhouse.GetTableSchema results when they contain credentials, connection strings, or internal implementation details.
  • Deny Clickhouse.DiscoverTables results that expose table names matching your secrets, payroll, legal-hold, or customer-data patterns, so the agent cannot use metadata to locate restricted datasets.

How to use Clickhouse with your agent

Choose your MCP client below and follow the setup to connect Clickhouse, configure its shared API key, and let your agent discover and query the database.

Your questions, answered

A ClickHouse MCP server exposes ClickHouse database capabilities as structured tools that an AI agent can call through the Model Context Protocol. An agent can discover databases, schemas, tables, table structures, and run read-only queries.

ClickHouse does not provide the MCP server used here. Arcade hosts the integration and adds the runtime around it, including authentication, approval policies, response redaction, and access from supported clients.

Arcade agents can discover databases, schemas, and tables; inspect a table's schema; and execute read-only ClickHouse SELECT queries. They cannot insert, update, delete, or change ClickHouse configuration through these actions.

ClickHouse requires a shared API key or secret with permission to access the databases and run the read-only queries your workflow needs. Grant that credential only the databases, schemas, tables, and query privileges the agent should reach.

No. This integration uses one shared API key or secret configured once, rather than having each user connect an individual ClickHouse account. Arcade brokers the credential, and the model never receives it.

Yes. Contextual Access policies can require human approval before a ClickHouse call, including a query that your organization treats as sensitive or expensive. The call proceeds only after the policy's conditions are satisfied; the available actions themselves are read-only.

Arcade proxies ClickHouse API calls and does not persist your database content as part of normal tool execution. The shared credential is brokered and stored encrypted at rest, and the model never receives it.

Any framework or client that can connect to Arcade's MCP gateway or tool APIs works, including LangChain, OpenAI Agents, CrewAI, Claude Desktop, Claude Code, Cursor, Visual Studio Code, Microsoft Copilot Studio, and GitHub Copilot.

Arcade refreshes access while the shared credential remains valid. If the key or secret is revoked, rotated, or can no longer be used, ClickHouse calls stop until the credential is reconnected or replaced.

Queries are subject to the limits configured on your ClickHouse deployment and any limits attached to the shared credential, including resource, concurrency, or query-duration limits. Arcade does not remove those ClickHouse-side limits.

Ready to actually ship your agent?

Connect your agent to Clickhouse with a shared API key and approval controls for consequential actions.

Build Your Agent Read the Clickhouse Quickstart