Show Sample Data
Paste a few real rows: samples beat descriptions.
Give Claude a few rows of real or realistic data so it can infer the shape, spot edge cases, and validate its own output. Describing a CSV is good; pasting three representative rows is better. Samples surface the messy reality that descriptions gloss over: inconsistent date formats, empty fields, extra whitespace. A sample is worth a thousand descriptions.
Where you'll practice this
One Promptles scenario teaches this principle directly.
The File Renamer
You have a folder called `deliverables/` with 500 client PDF files named inconsistently: some are `ACME_project123_Mar2025.pdf`, others are `project123-acme-2025-03.pdf`, and some are just `acme.pdf`. You want a Python script called `rename.py` that renames every PDF to the…
Other principles in Context
Declare the Tools You Need
Name the tool or MCP server by name, and say what to do if it isn't reachable.
Describe Before You Prescribe
Describe the Data Shape
Claude can't clean data it can't picture.
Explain the Integration Point
Keep Memory Reusable
Save patterns and preferences that survive across sessions; never save snapshots of finished work.
Map to Real Components
Use Code Connect so it reuses your components, not new markup.