← 返回博客列表
2026-02-20来自 LinkedIn

cwprep v0.3.0 Release: Introducing the Text-to-PrepFlow Engine

cwprep v0.3.0 Release: Introducing the Text-to-PrepFlow Engine Text-to etl for tableau in 2 steps: No install, no code, no drag. Two weeks ago, I open-sourced cwprep - a Python SDK that enables programmatic...

linkedinimported

原始来源: LinkedIn

Original LinkedIn Post

cwprep v0.3.0 Release: Introducing the Text-to-PrepFlow Engine

Text-to etl for tableau in 2 steps: No install, no code, no drag.

Two weeks ago, I open-sourced cwprep - a Python SDK that enables programmatic generation of Tableau Prep data flows by reverse-engineering the native .tfl JSON structure.

Today, with the release of v0.3.0, the project evolves into a true Text-to-PrepFlow Engine. Powered by the Model Context Protocol (MCP), cwprep now acts as a direct bridge between Large Language Models (like Claude and Gemini) and Tableau Prep. This allows data engineers and analysts to generate production-ready data cleaning flows entirely through conversational AI.

Key updates in v0.3.0: Enhanced Data Cleaning: Added 8 new quick calculation operations (trimming, case conversion, etc.), column data type transformation, and column duplication.

Expanded Operator Coverage: When compared against the published capabilities of the upcoming official Tableau Agent, our operator support has now reached 88% coverage, while exclusively supporting advanced operations like cross-component joins and unions. Our goal is to bring AI-driven automation to traditional visual data flow tools.

#TableauPrep #DataAnalytics #AIAutomation #DataEngineering #OpenSource #gemini #google #claudecode #ai #prep #text2etl #datafam #