Prompt Efficiency

A precise prompt is a fast prompt. Not because Claude processes it faster — processing time scales with output, not input — but because a precise prompt reduces the number of iterations needed to get the right result. "Add error handling to the auth module" requires two to three follow-up clarifications. "Add try-catch blocks to all async functions in src/auth/validate.ts. Catch errors should log the error with console.error and re-throw." requires zero follow-ups.

The token cost of imprecise prompts is multiplicative. Each clarification cycle costs the original prompt tokens (re-read by Claude as context), the clarification question tokens, your response tokens, and the re-attempt tokens. A prompt that needs three clarification cycles costs roughly four times what a precise prompt costs. The time cost is even worse — each cycle takes minutes, and three cycles turn a two-minute task into a fifteen-minute task.

# Inefficient (3 rounds, ~4x token cost):
"Fix the auth module"
→ "Which file?" → "validate.ts"
→ "What kind of fix?" → "Error handling"
→ "What error handling pattern?" → "try-catch, log, re-throw"

# Efficient (1 round, 1x token cost):
"Add try-catch to all async functions in src/auth/validate.ts.
Catch blocks: console.error(error), then re-throw.
Do not change function signatures."