prompt-optimizer
Transform vague prompts into precise, well-structured specifications using EARS (Easy Approach to Requirements Syntax) methodology. This skill should be used when users provide loose requirements, ambiguous feature descriptions, or need to enhance prompts for AI-generated code, products, or documents. Triggers include requests to "optimize my prompt", "improve this requirement", "make this more specific", or when raw requirements lack detail and structure.
88
83%
Does it follow best practices?
Impact
96%
1.62xAverage score across 3 eval scenarios
Passed
No known issues
Evaluation results
98%
11%Notification Center Requirements
EARS syntax transformation
Issues identified
100%
100%
Event-driven pattern used
70%
100%
State-driven pattern used
62%
100%
Conditional pattern used
62%
100%
Unwanted behavior pattern used
62%
100%
Measurable criteria present
100%
100%
Vague language removed
100%
100%
Atomic statements
87%
75%
Implicit conditions made explicit
100%
100%
Domain theories cited
100%
100%
EARS verbs used
100%
100%
How-to-use guidance
100%
100%
97%
50%Loyalty Program App Requirements Enhancement
Enhanced prompt structure and domain grounding
Role section present
0%
100%
Role specificity
80%
100%
Skills section present
0%
100%
Skills count
66%
100%
Workflows section present
0%
100%
Workflows completeness
50%
100%
Examples section present
0%
100%
Examples use real data
0%
100%
Formats section present
0%
100%
Formats specificity
30%
100%
Domain theories cited
80%
70%
Theories applied to features
100%
100%
How-to-use guidance
100%
100%
94%
51%Improving a Vague SaaS Onboarding Requirement
Six-step workflow and structured output format
Original requirement included
100%
100%
Issues section present
88%
100%
Issues are specific
100%
100%
EARS transformation section
0%
100%
Trigger-based requirement
0%
100%
State-based requirement
0%
100%
Prevention requirement
0%
100%
Domain section present
0%
100%
Named theories with relevance
0%
100%
Enhanced prompt section present
62%
100%
Concrete examples in prompt
87%
100%
Measurable formats in prompt
75%
100%
How-to-use guidance
33%
33%
- Repository
- fernandezbaptiste/claude-code-skills
- Commit
4f0eae8
- 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.