promptify
Transform user requests into detailed, precise prompts for AI models. Use when users say 'promptify', 'promptify this', or explicitly request prompt engineering or improvement of their request for better AI responses.
89
Quality
87%
Does it follow best practices?
Impact
89%
1.58xAverage score across 3 eval scenarios
Evaluation results
94%
52%Improving a Vague Documentation Request
Four-block structure and measurable specificity
Output file exists
100%
100%
Labeled markdown block
25%
25%
Context block present
50%
100%
Task block present
37%
100%
Constraints block present
12%
100%
Output Format block present
25%
100%
No vague qualifiers
40%
100%
Audience specified
100%
100%
No emojis
100%
100%
No asterisk bullets
100%
100%
No first-level heading
0%
100%
Two follow-up options
20%
100%
Save-to-file naming convention
0%
100%
Without context: $0.1062 · 34s · 9 turns · 14 in / 1,634 out tokens
With context: $0.3625 · 1m 13s · 23 turns · 2,664 in / 3,616 out tokens
87%
31%Refining a Multi-Step Data Pipeline Request
Complex task decomposition and output format specification
promptified.md exists
100%
100%
Epoch-seconds filename
0%
100%
Numbered sequential steps
100%
100%
Output format specified
100%
100%
No vague qualifiers
40%
100%
Specification language
80%
100%
Labeled prompt block
0%
0%
No emojis
100%
100%
No asterisk bullets
100%
100%
Two follow-up options
0%
100%
No invented information
20%
50%
Without context: $0.3477 · 1m 38s · 21 turns · 26 in / 4,826 out tokens
With context: $0.4637 · 1m 46s · 27 turns · 1,487 in / 5,932 out tokens
87%
17%Cleaning Up a Chatty AI Request for Code Review
Specification language, writing style, and length constraint
promptified.md exists
100%
100%
No emojis in output
100%
100%
No filler phrases
100%
100%
Specification language
100%
100%
Length proportional
100%
100%
No first-level heading
0%
100%
No asterisk bullets
100%
100%
Opening summary paragraph
60%
100%
Labeled prompt block
0%
0%
Two follow-up options
33%
100%
No invented content
50%
37%
Without context: $0.1184 · 28s · 8 turns · 11 in / 1,294 out tokens
With context: $0.1991 · 40s · 15 turns · 1,318 in / 2,056 out tokens
- Evaluated
- Agent
- Claude Code
- Model
- Claude Sonnet 4.6
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.