Codepit The toolbox

Claude follows ClickHouse's own best practices

Official ClickHouse skills for schema design, query tuning and data ingestion, plus chdb for running ClickHouse SQL inside Python. Claude checks its advice against the rules and names the rule it used.

Scan in Codepit View source

Scan checks all of ClickHouse/agent-skills on your computer, before anything installs.

Codepit runs on your computer. Open this page there to scan it before installing.

An open manual with a honey ribbon

What you get

Eleven skills in all, among them:

  • clickhouse-best-practices
  • clickhouse-architecture-advisor
  • clickhouse-js-node-troubleshooting
  • chdb-datastore
  • chdb-sql
  • infra-clickhouse
  • infra-postgres
  • clickstack-otel-collector

When it kicks in

  • CREATE TABLE
  • ORDER BY keys
  • Data types
  • Slow queries
  • JOINs
  • Ingestion pipelines
  • Updates and deletes
  • Querying files with chdb

Each skill kicks in on its own when your work touches it.

Before you start

  • For the two chdb skills: Python 3.9 or later on macOS or Linux, and pip install chdb.
  • For infra-clickhouse, infra-postgres and the ClickStack skill: the clickhousectl CLI. The skills check for it and show how to install it.

Try saying

>Create a table for storing user events with user_id, event_type, JSON properties and a timestamp

Claude reads the matching rules, designs the table and cites each rule it applied.

Worth knowing

The infra skills can create ClickHouse Cloud services, which cost money. They are told to default to a local setup and to ask you before creating anything in the cloud.

The ClickStack skill creates a dedicated ingest SQL user on your ClickHouse Cloud service and sends it synthetic logs, traces and metrics to check the pipeline end to end.

Managed ClickHouse Cloud Postgres, which infra-postgres covers, is in beta.

Read the author's README README.md