| name | clickhouse |
| description | Connect to and query ClickHouse (a local server or a ClickHouse Cloud service) from the terminal using the official clickhousectl CLI, including the browser OAuth login flow. Use when the user wants to run SQL against ClickHouse, explore schemas and tables, inspect Cloud services, or authenticate clickhousectl. For building a local dev environment or deploying to Cloud, defer to the official ClickHouse skills (see Scope). |
ClickHouse via clickhousectl
Connect to ClickHouse and run queries using clickhousectl, the official ClickHouse CLI. This skill covers the parts a data analyst needs: authenticating, pointing at the right server, and running safe SQL. It does not use the ClickHouse MCP server; everything goes through the CLI.
Scope
This skill is for connecting and querying. For these other flows, use the bundled official ClickHouse skills (siblings in this directory) instead of reinventing them:
- Setting up a local dev environment from scratch (install ClickHouse, init a project, start a server, create schema):
../clickhousectl-local-dev/.
- Deploying to or migrating into ClickHouse Cloud (create a service, migrate schema, provision an app user):
../clickhousectl-cloud-deploy/.
- Writing or optimizing non-trivial SQL, or the agent schema-discovery and query-safety workflow:
../clickhouse-best-practices/.
- Running SQL on local files or remote sources without a server:
../chdb-sql/.
These are vendored from https://github.com/ClickHouse/agent-skills (Apache-2.0). They can also be installed standalone with clickhousectl skills or npx skills add clickhouse/agent-skills.
Step 1: Ensure clickhousectl is installed
which clickhousectl
If not found, install it (downloads the right build for the OS, installs to , and creates a alias):
