← 返回博客列表
2026-02-23来自 LinkedIn

Just dropped: 8 ready-to-use prompt templates for cwprep MCP

Just dropped: 8 ready-to-use prompt templates for cwprep MCP What if you could build a Tableau Prep data flow just by?describing your business need? With?cwprep, you literally just say: "Find loss-making orders with...

linkedinimported

原始来源: LinkedIn

Original LinkedIn Post

Just dropped: 8 ready-to-use prompt templates for cwprep MCP

What if you could build a Tableau Prep data flow just by?describing your business need?

With?cwprep, you literally just say: "Find loss-making orders with excessive shipping duration"

gemini cli's answer: To achieve this, I will build the process according to the following logic: ??1. Read Data: Read order data from MySQL's orders table. ??2. Calculate Shipment Time: Add the Calculation field shipping_duration to calculate the difference in days between ship_date and order_date. ??3. Filter Losing Orders: Filter out orders with a profit < 0. 4. Filter Long Orders: Filter out orders with shipping_duration > 5 (assuming more than 5 days are too long). ??5. Output Results: Save the analysis results as a TFL file.

And the AI generates a complete?.tfl?file ? with joins, calculated fields, filters, and output ? all through a single MCP conversation. No GUI clicking. No JSON hacking. I've just published?8 real-world prompt templates?covering:

Sales analysis & regional reports Return rate monitoring Logistics efficiency warnings VIP customer segmentation

Each prompt comes in 2 flavors: Full Prompt?? precise control over every step Capability Test?? let the AI figure out the logic itself

The best part? Setup is literally one line in claude code: claude mcp add cwprep -- uvx --from "cwprep[mcp]" cwprep-mcp

Then plug it into Claude, Cursor, Gemini, VS Code? done. See more in github.

Prompts here:? examples/prompts md

Text ? TableauPrepFlow. That's it. Try it now!

#Tableau #MCP #AI #DataEngineering #OpenSource #LLM #cwprep #text2prep #tableauprep #gemini #claudecode #automation #datafam