skills.rest
JNZader-Vault
JNZader-VaultOFFICIAL
Explore all skills by this author →
GitHub

prompt-improver

Analyze, research, question, and execute ambiguous user prompts.

Updated Feb 2, 2026
Search Tags:#workflow#prompt engineering#clarification#ai assistant#contextualization#prompt refinement
One-click install
npx skills add https://github.com/JNZader-Vault/project-starter-framework --skill prompt-improver-jnzader-vault
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: prompt-improver
Source: https://github.com/JNZader-Vault/project-starter-framework/tree/main/.ai-config/skills/prompt-improver
Command: npx skills add https://github.com/JNZader-Vault/project-starter-framework --skill prompt-improver-jnzader-vault

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill tackles the issue of vague or unclear user prompts by systematically enriching them with context before execution, ensuring more accurate and relevant AI responses.

Core Features & Use Cases

  • Vagueness Detection: Automatically identifies prompts lacking clarity, action verbs, or context.
  • Contextual Research: Gathers relevant information from the codebase to inform clarifying questions.
  • Targeted Questioning: Asks precise, actionable questions to the user to resolve ambiguity.
  • Enriched Execution: Executes tasks with a full understanding of user intent and project context.
  • Use Case: When a user asks "Make this better," the skill will analyze the request, search for related code, ask specific questions like "Should this refactor use the existing AuthService or a new one?", and then proceed with the improved understanding.

Quick Start

Use the prompt improver skill to clarify the request "update the user profile".

Frequently Asked Questions about prompt-improver

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

FAQPage Schema
How do I refine vague AI prompts for better code generation?▼

To refine vague AI prompts, this skill detects missing action verbs and context, searches the codebase for relevant information, and asks targeted clarifying questions to enrich execution.

What is the best way to clarify ambiguous requests in an AI assistant workflow?▼

The best way to clarify ambiguous requests is using a structured four-phase process: Analyze, Research, Question, and Execute, which systematically gathers project context before proceeding.

How does contextualization improve AI-driven task execution in software development?▼

Contextualization improves AI-driven task execution by gathering relevant codebase information beforehand, allowing the assistant to ask precise questions and resolve vagueness before generating code.

Can I use this skill to fix a prompt lacking clear action verbs and context?▼

Yes, you can use this skill to fix a prompt lacking clear action verbs because it automatically detects vagueness and prompts you with specific questions to resolve the missing context.

What happens when you ask an AI assistant to 'make this better' without context?▼

When you ask an AI assistant to 'make this better' without context, this skill identifies the vagueness, searches your code for related dependencies like AuthService, and asks you to clarify the desired refactoring approach.

Do I need to provide context manually when prompt refinement detects vagueness?▼

No, you do not need to provide context manually because the skill automatically conducts code searches to gather relevant project information before asking you targeted clarifying questions.

Related Skills

Explore More Software Engineering Skills→
Fr-e-dFr-e-d
157

architecture-extract

Analyze project directory structures to generate architectural summaries.

Community
Basic
shogoggshogogg
4

coderabbit-review

Automate CodeRabbit code reviews for uncommitted and committed Git changes.

Community
Advanced
VitaDynamicsVitaDynamics

ask

Answer codebase questions by orchestrating parallel exploration agents.

Official
Intermediate
deerawandeerawan
1

gsd-spec-phase

Run structured Socratic questioning with ambiguity scoring to write SPEC.md.

Community
Advanced