enhance-prompt

Rewrites vague or incomplete prompts into structured, self-contained enriched prompts.

1|Updated Apr 20, 2026
One-click install
npx skills add https://github.com/ntd237/draft_toolkits_20042026 --skill enhance-prompt-ntd237
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: enhance-prompt
Source: https://github.com/ntd237/draft_toolkits_20042026/tree/main/.agents/skills/enhance-prompt
Command: npx skills add https://github.com/ntd237/draft_toolkits_20042026 --skill enhance-prompt-ntd237

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Raw prompts typed into AI assistants are often missing role, context, output format, or constraints, which leads to generic or wrong answers and wasted iterations. This Skill detects those gaps automatically and rewrites the input into one complete, ready-to-use prompt before it is executed or passed to another skill. ## Core Features & Use Cases - Automatic Gap Detection: Scores every prompt across 7 dimensions (role, task, context, input, output, constraints, goal) plus 7 programming-specific checks (language, framework, environment, error message, codebase scope, performance, integrations). - Infer-or-Ask Decision Logic: Infers low-risk gaps with smart defaults and asks at most 2 targeted multiple-choice questions for high-risk gaps like target framework or production vs. prototype quality. - Structured Enriched Output: Produces exactly one self-contained prompt with Role, Task, Input, Expected Output, Constraints, Success Criteria, and an honest assumptions footer. - Use Case: A user types "help me fix my API". The Skill detects missing language, framework, and error message, asks one clarifying question, then outputs a complete debugging prompt specifying FastAPI, Python 3.11, the traceback placeholder, and expected vs. actual behavior. ## Quick Start Paste your rough prompt and ask the assistant to enhance it into a complete, ready-to-use prompt before running it.

Frequently Asked Questions about enhance-prompt

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I improve a vague prompt before sending it to an AI?▼

Run the prompt through an enrichment pass that checks role, task, context, input, output format, constraints, and goal. Missing low-risk details are inferred with stated assumptions, while high-risk gaps like target framework trigger a short clarifying question.

What makes a good prompt for programming tasks?▼

A strong programming prompt specifies the language and version, framework, environment, exact error message or traceback, codebase scope, and expected output format. It also states whether the code is prototype or production quality and lists constraints and success criteria.

When should the enhancer ask questions instead of assuming?▼

It asks only for high-risk gaps where a wrong assumption wastes significant effort, such as target language or framework, modifying existing code versus writing from scratch, and production versus prototype requirements. It asks at most two questions total.

Does prompt enrichment work for debugging and AI/ML tasks?▼

Yes. Debugging prompts get a checklist for error message, minimal reproducible code, expected versus actual behavior, and prior attempts. ML prompts get fields for model architecture, dataset size, training hardware, and performance baseline.

What are the limitations of automatic prompt enrichment?▼

It rewrites prompts but never executes or answers them, and it avoids inventing technical details like version numbers, using placeholders instead. Output is capped at 200 lines and limited to a single prompt variant.