ai-docs/docs/ai/crossplatform/prompting-antipatterns.md

Prompting Anti-Patterns (And Fixes)

You are here: AI Docs Home > Cross-Platform AI Usage > Prompting Anti-Patterns

Contents

• Why Anti-Patterns Cause Rework
• Common Anti-Patterns and Fixes
• Quick Self-Check Before Sending
• Next Steps

Why Anti-Patterns Cause Rework

Anti-patterns are prompt styles that are too broad or ambiguous. They usually produce low-quality output, extra retries, and higher token use.

Common Anti-Patterns and Fixes

Too broad:

Fix everything in this project.

Better: Example prompt:

Refactor this file to remove duplication. Do not change behavior.

Too vague:

Make this code better.

Better: Example prompt:

Improve naming and add comments only where logic is complex.

Missing inputs:

Update the service to handle retries.

Better: Example prompt:

Update ServiceA in src/service/ServiceA.kt to retry 2 times on 5xx. Keep API the same.

Quick Self-Check Before Sending

1. Did I name the exact file or function?
2. Did I say what must not change?
3. Did I ask for one clear output type?

Next Steps

• Read Prompting Patterns for reusable templates.
• Use Plan-First Workflow for larger changes.