The TypeScript Semantic Layer for ClickHouse

Name: The TypeScript Semantic Layer for ClickHouse
Availability: InStock
Author: lureilly1

by lureilly1·Jun 27, 2026·2 points·1 comment

Visit Project View on HN

AI Analysis

●●SolidSolve My ProblemBig Brain

Schema-driven types and MCP server, but semantic layers like Cube already exist.

Strengths

•Live schema introspection generates TypeScript types automatically, preventing drift
•MCP server integration lets LLM agents query typed datasets safely
•No YAML or separate platform—runs as a library where your app runs

Weaknesses

•ClickHouse-only limits audience compared to database-agnostic alternatives
•Semantic layer category is crowded with Cube, dbt Metrics, and Transform

Post Description

I've built a type-safe semantic layer in code, for ClickHouse. If you're building analytics off ClickHouse in TypeScript, I would love your feedback.

With hypequery there is no platform to adopt, no YAML sprawl. It runs where your app runs.

Key features:

- Define metrics once, reuse them everywhere: Declare dimensions and measures in one place and then pull from the same source of truth.

- Compiles to ClickHouse SQL: No service, no proxy, no extra runtime to deploy. It's a library that generates SQL and runs where your app runs.

- Multi-tenancy & Authentication ready: Cross-tenant queries are blocked at the query layer, helpers to plug into your existing auth.

- Agent-native: A dataset is a declared set of dimensions and measures, so it doubles as an allowlist. Includes an MCP server to hand an LLM a typed catalog to query.

- Runtime HTTP entry point: serve() exposes any dataset as an endpoint, so the same type-safe definitions back your dashboards and your API.