# Refine Prompts With Structured Analysis

Reusable AI prompts often fail because goals, context, and output requirements are unclear. This skill helps you analyze, restructure, and improve prompts with systematic checks.

## Install

```bash
npx skillstore add dnyoussef/prompt-architect
```

## Metadata

- Status: approved
- Slug: dnyoussef-prompt-architect
- Skillstore revision: r1
- Version status: missing
- Tree hash: 942a43bad76688f699f61bc635bd7a2ec176f2292e1e9249dffb7b601e13f95b
- Author: DNYoussef
- GitHub username: DNYoussef
- License: MIT
- Repository: https://github.com/DNYoussef/ai-chrome-extension/tree/main/.claude/skills/prompt-architect
- Ref: 72d5025b022c77f7a51bdf5c1637c689c80e89d1
- Supported tools: Claude, Codex, Claude Code
- Audit status: complete
- Agent install advisory: allowed
- Manual install advisory: allowed
- Artifact signature: available
- Audit attestation: unavailable
- Human verification: not\_verified
- Risk factors: external\_commands
- Quality score: 74
- Public page: https://skillstore.pages.dev/skills/dnyoussef-prompt-architect
- Manifest: https://skillstore.pages.dev/api/skills/dnyoussef-prompt-architect/manifest

## Capabilities

- Reviews prompts for clarity, structure, context, and output requirements.
- Identifies common prompt anti-patterns such as vague goals, contradictions, and missing edge cases.
- Suggests evidence-based techniques including few-shot examples, self-consistency checks, and plan-and-solve structure.
- Adapts prompt guidance for analytical, coding, creative writing, transformation, and question-answering tasks.
- Helps convert weak prompts into reusable templates for teams or repeated workflows.

## Use Cases

- Improve Team Prompt Templates: Convert inconsistent team prompts into clearer reusable templates with explicit context, constraints, and output expectations.
- Design Reliable Coding Prompts: Create prompts that specify language, version, requirements, error handling, and expected behavior for code generation tasks.
- Teach Prompt Engineering Basics: Use the framework to explain why prompt changes improve clarity, reliability, and maintainability.

## Prompt Templates

### Improve a Simple Prompt

```
Review this prompt for clarity and missing context. Explain the main weaknesses, then provide a revised version that keeps the original goal.
```

### Create a Reusable Template

```
Turn this task into a reusable prompt template. Include placeholders, required context, output format, edge cases, and quality checks.
```

### Audit a Prompt Library

```
Analyze these prompts as a library. Group recurring issues, identify duplicated patterns, recommend standard sections, and prioritize fixes by impact.
```

### Design a High-Stakes Workflow Prompt

```
Design a prompt for this high-stakes workflow. Include planning, execution, verification, uncertainty handling, and clear limits on what the model should not assume.
```

## Limitations

- It provides prompt design guidance, not guaranteed model performance.
- It does not test prompts against live model outputs unless the user provides results.
- It cannot validate facts, citations, or domain claims without source material.
- It may need user context to tailor prompts for specialized industries or policies.

## Best Practices

- State the task, audience, constraints, and desired output before adding examples.
- Use examples to clarify expected behavior, especially for repeated or team workflows.
- Test refined prompts with edge cases and revise based on observed failures.

## Anti Patterns

- Asking for broad analysis without defining purpose, scope, or success criteria.
- Adding many nested rules before the core task is clear.
- Assuming the model knows private standards, team formats, or missing background.

## Security Audit

- Audited at: 2026-07-06T11:21:48.013\+00:00
- Summary: All twelve static detections are false positives caused by prose examples and Markdown delimiters. No evidence found of command execution, network reconnaissance, system reconnaissance, prompt injection, or data-exfiltration intent in the reviewed files.

## Stats

- Views: 200
- Downloads: 7
- Favorites: 0
- Popularity score: 0