4 phút đọc
dbt-labs/dbt-charts: Declarative YAML around SQL for dbt-native dashboards: easy for agents to write, easy for – Repository dbt-labs/dbt-charts. Mô tả: Declarative YAML around SQL for dbt-native dashboards: easy for agents to write, easy for humans to audit.. 563 stars, 18 forks, ngôn ngữ chính Python, cập nhật 2026-10-12.
1. Repository làm gì?
A declarative YAML syntax around SQL for making dashboards.
dbt Charts, from dbt Labs (package dbt-charts, CLI dct), compiles YAML board definitions into interactive dashboards and reports in many formats. SQL is still the language for defining WHAT data you want to see, and dbt Charts wraps that with a simple YAML syntax declaring HOW you want to see it.
Problem: Vibe coded dashboards create a mess of artifacts, frameworks and data transformations at every layer, making it hard or impossible to audit.
Solution: a clear YAML syntax for dashboards that's easy for agents to write and humans to audit.
Run this, then follow what it prints:
No shell? Read https://docs.dbtcharts.com/llms.txt and the MCP page. The whole reference in one fetch: https://docs.dbtcharts.com/llms-full.txt.

2. Dữ liệu và cấu trúc đáng chú ý
Try it without installing anything at play.dbtcharts.com, or start with the docs.
This repository is a read-only mirror of dbt Labs' private upstream. Issues are welcome; pull requests are not accepted. See CONTRIBUTING.md.
Beta. dbt Charts is pre-1.0. Requires Python 3.11 to 3.13.
Hand your coding agent one sentence. It runs the command in Are you an AI agent? and follows what it prints:
dct talks to your warehouse through dbt adapters and needs the adapter in its own environment, so install the one you use as an extra:
Unreleased main ships as an unsupported dev build on every export:
dct serve renders every board under charts/ and lets you move between them from the breadcrumb:

3. Khả năng ứng dụng và giới hạn
That is the short path. dct is a full CLI: it validates, renders to HTML, PDF, and PNG, runs queries, searches boards, traces column impact, and carries the YAML reference offline. The dct CLI section below lists the verbs.
Explore the dbtcharts.com platform for hosted boards and an enhanced collaboration experience.
While in beta, dbt Charts ships as a separate CLI, dct, short for data chart tool. It mirrors dbt's shape: one verb per job.
Project verbs read DCT_PROJECT_DIR when –project-dir is not passed. The CLI reference lists every environment variable.
The dct CLI ships with all the skills and tools agents need to query datasets and write, render, and serve dashboards. It is the recommended tool for agents to work with, but also ships with a local MCP server for surfaces like VS Code where one is required.
dbt Charts for VS Code adds syntax highlighting for the YAML, the SQL inside it, and Jinja; schema-backed autocomplete and hover docs; diagnostics as you type; and a live preview beside the editor that re-renders on save and honors the board's filters. The preview shells out to dct render, so install the CLI first. Details in the docs.
dbt Charts is also available on Open VSX for VS Code-compatible editors (e.g. Cursor, Windsurf)
One variable, one query, one chart, and the whole board fits on a screen. It runs against the Playground's demo data, and the link under it opens the exact YAML there, live and editable:
Open this board in the Playground
Số liệu repository có thể thay đổi; nên mở liên kết nguồn để kiểm tra README, giấy phép và trạng thái phát hành.
Nguồn tham khảo: Xem bài gốc