Prompt Framework Selection Guide

中文版本

📖 Introduction

This guide helps you choose the most suitable prompt framework based on different testing scenarios and requirements. We provide 6 prompt frameworks, each with unique advantages and applicable scenarios.

🎯 Quick Selection Flowchart

graph TD
    A[Start Selection] --> B{New User?}
    B -->|Yes| C[Standard Framework]
    B -->|No| D{Need Deep Analysis?}
    D -->|Yes| E[CRISPE Framework]
    D -->|No| F{Emphasize Structure?}
    F -->|Yes| G[LangGPT Framework]
    F -->|No| H{Need Quick Start?}
    H -->|Yes| I[RISE Framework]
    H -->|No| J{Emphasize Scenario & Steps?}
    J -->|Yes| K[ROSES Framework]
    J -->|No| L[ICIO Framework]

📊 Framework Comparison Table

Framework	Complexity	Learning Curve	Output Quality	Use Cases	Recommendation
Standard	⭐	Easy	⭐⭐⭐⭐	General	⭐⭐⭐⭐⭐
ROSES	⭐⭐⭐	Medium	⭐⭐⭐⭐⭐	Complex Projects	⭐⭐⭐⭐⭐
LangGPT	⭐⭐⭐⭐	Hard	⭐⭐⭐⭐⭐	Systematic	⭐⭐⭐⭐
ICIO	⭐⭐	Easy	⭐⭐⭐⭐	Data-Driven	⭐⭐⭐⭐
CRISPE	⭐⭐⭐	Medium	⭐⭐⭐⭐⭐	Deep Analysis	⭐⭐⭐⭐⭐
RISE	⭐⭐	Easy	⭐⭐⭐⭐	Quick Tasks	⭐⭐⭐⭐

🔍 Detailed Framework Selection Guide

1. Standard Framework - Universal Choice

Suitable For:

• ✅ First-time AI-assisted testing users
• ✅ Teams needing quick onboarding
• ✅ Users unfamiliar with framework structures

Core Structure:

Role: Role definition
Context: Background context
Task: Specific task

Advantages:

• Simple and intuitive structure
• Low learning cost
• Wide applicability

Usage Recommendations:

• Suitable for daily testing tasks
• Recommended as entry-level framework
• Quick test plan generation

Example Scenarios:

• API test plan design
• Test case writing
• Bug report generation

---

2. ROSES Framework - Best for Complex Projects

Suitable For:

• ✅ Projects requiring detailed step guidance
• ✅ Complex testing scenarios
• ✅ Tasks needing clear roles and scenarios

Core Structure:

Role: Role
Objective: Objective
Scenario: Scenario
Expected Solution: Expected solution
Steps: Steps

Advantages:

• Most complete framework
• Covers role, objective, scenario, solution, and steps
• Suitable for complex tasks requiring detailed guidance

Usage Recommendations:

• Suitable for large testing projects
• Tasks requiring phased execution
• Scenarios demanding detailed steps

Example Scenarios:

• Complete test strategy formulation
• Performance testing for complex systems
• Multi-phase security testing plans

---

3. LangGPT Framework - Systematic Expert Choice

Suitable For:

• ✅ Highly structured projects
• ✅ Pursuing maintainability and reusability
• ✅ Test engineers with programming mindset

Core Structure:

# Role: Role name
## Profile:
- author: Author
- version: Version
- language: Language
- description: Description

## Goals: Goals
## Constrains: Constraints
## Skills: Skills
## Workflows: Workflows
## Initialization: Initialization

Advantages:

• Most systematic framework
• Template design, easy to maintain
• Supports variables and modularization

Usage Recommendations:

• Suitable for long-term maintained testing frameworks
• Team collaboration projects
• Scenarios requiring version control

Example Scenarios:

• Enterprise-level test system construction
• Reusable test template development
• Standardized test process design

---

4. ICIO Framework - Data-Driven Preferred

Suitable For:

• ✅ Scenarios emphasizing input/output
• ✅ Need clear evaluation metrics
• ✅ Data-driven testing projects

Core Structure:

Instruction: Instruction
Context: Context
Input Data: Input data
Output Indicator: Output indicator

Advantages:

• Emphasizes data flow
• Clear input/output
• Easy to quantify evaluation

Usage Recommendations:

• Suitable for API testing
• Data validation scenarios
• Tests requiring clear metrics

Example Scenarios:

• API interface testing
• Database testing
• Performance metric validation

---

5. CRISPE Framework - Deep Analysis Preferred

Suitable For:

• ✅ Scenarios requiring in-depth analysis
• ✅ Clear output style requirements
• ✅ Need multiple example references

Core Structure:

Capacity and Role: Capacity and role
Insight: Insight
Statement: Statement
Personality: Personality
Experiment: Experiment

Advantages:

• Emphasizes capability and insight
• Supports personalized output
• Encourages diverse examples

Usage Recommendations:

• Suitable for tests requiring deep analysis
• Specific style output requirements
• Need multiple solution comparisons

Example Scenarios:

• Security testing deep analysis
• Performance bottleneck diagnosis
• Multi-solution test strategy design

---

6. RISE Framework - Quick Task Tool

Suitable For:

• ✅ Tasks needing quick completion
• ✅ Simple and clear testing scenarios
• ✅ Time-critical projects

Core Structure:

Role: Role
Input: Input
Steps: Steps
Expectation: Expectation

Advantages:

• Most concise framework
• Quick to start
• High execution efficiency

Usage Recommendations:

• Suitable for urgent tasks
• Simple testing scenarios
• Quick idea validation

Example Scenarios:

• Quick bug verification
• Simple functional testing
• Ad-hoc testing tasks

---

🎨 Scenario-Based Selection Recommendations

By Testing Type

API Testing

1. Preferred: ICIO Framework (emphasizes input/output)
2. Alternative: Standard Framework (general scenarios)
3. Deep: CRISPE Framework (detailed analysis needed)

Performance Testing

1. Preferred: ROSES Framework (detailed steps needed)
2. Alternative: LangGPT Framework (systematic approach)
3. Quick: RISE Framework (quick validation)

Security Testing

1. Preferred: CRISPE Framework (deep analysis)
2. Alternative: ROSES Framework (complete process)
3. Systematic: LangGPT Framework (standardized)

Functional Testing

1. Preferred: Standard Framework (general scenarios)
2. Alternative: ROSES Framework (complex features)
3. Quick: RISE Framework (simple features)

Test Case Writing

1. Preferred: ROSES Framework (detailed steps)
2. Alternative: Standard Framework (regular cases)
3. Systematic: LangGPT Framework (templated)

By Project Scale

Small Projects (< 1 month)

• Recommended: Standard, RISE
• Reason: Quick start, efficiency priority

Medium Projects (1-6 months)

• Recommended: ROSES, ICIO
• Reason: Balance complexity and completeness

Large Projects (> 6 months)

• Recommended: LangGPT, CRISPE
• Reason: Systematic, strong maintainability

By Team Experience

Junior Teams

• Recommended: Standard → RISE → ICIO
• Learning Path: Simple to complex, progressive

Intermediate Teams

• Recommended: ROSES → ICIO → CRISPE
• Learning Path: Focus on practicality and depth

Senior Teams

• Recommended: LangGPT → CRISPE → Custom
• Learning Path: Systematization and customization

---

💡 Usage Tips

1. Hybrid Usage Strategy

Different testing phases can use different frameworks:

• Requirements Analysis: Standard or ROSES
• Test Design: LangGPT or CRISPE
• Test Execution: ICIO or RISE
• Result Analysis: CRISPE or ROSES

2. Progressive Learning

Recommended learning order:

1. Standard (Foundation)
2. RISE (Simplified)
3. ICIO (Data)
4. ROSES (Complete)
5. CRISPE (Deep)
6. LangGPT (Systematic)

3. Version Selection Recommendations

Each framework has Full and Lite versions:

• Full Version: Suitable for complex projects, detailed output
• Lite Version: Suitable for quick tasks, concise output

Selection Recommendations:

• First-time use: Start with Lite version to familiarize with structure
• Formal projects: Use Full version for complete output
• Time-critical: Use Lite version for quick completion

---

📚 Practical Cases

Case 1: E-commerce Platform API Testing

Project Characteristics:

• Medium-sized project
• Numerous API interfaces
• Detailed test plan needed

Framework Choice: ICIO Framework

Rationale:

• Emphasizes input/output validation
• Suitable for API testing scenarios
• Facilitates data-driven testing

Case 2: Financial System Security Testing

Project Characteristics:

• Large-scale project
• High security requirements
• Deep analysis needed

Framework Choice: CRISPE Framework

Rationale:

• Supports deep security analysis
• Provides multiple testing approaches
• Emphasizes professionalism and insight

Case 3: Mobile App Functional Testing

Project Characteristics:

• Small project
• Relatively simple features
• Time-critical

Framework Choice: RISE Framework

Rationale:

• Quick to start
• Concise and efficient
• Meets basic testing needs

---

🔄 Framework Migration Guide

From Standard to ROSES

Applicable Scenario: Increased project complexity

Migration Steps:

1. Retain Role and Task sections
2. Expand Context into Scenario
3. Add Objective and Expected Solution
4. Detail Steps section

From RISE to CRISPE

Applicable Scenario: Need deeper analysis

Migration Steps:

1. Expand Role to Capacity and Role
2. Transform Input into Insight
3. Expand Expectation into Statement
4. Add Personality and Experiment

---

❓ FAQ

Q1: How to choose between Full and Lite versions?

A:

• Learning phase: Lite version
• Formal projects: Full version
• Quick tasks: Lite version
• Complex projects: Full version

Q2: Can I mix multiple frameworks?

A: Yes. Different testing phases can use different frameworks, but it's recommended to maintain framework consistency within the same phase.

Q3: Which framework is best for beginners?

A: Standard Framework. Simple structure, low learning cost, wide applicability.

Q4: How to evaluate if framework selection is appropriate?

A: Observe these indicators:

• AI output quality meets requirements
• Ease of use
• Team understanding
• Improved work efficiency

---

📖 Related Resources

• TOP 5 Frameworks Introduction
• Detailed Framework Documentation
• Testing Type Application Examples

---

Last Updated: 2026-01-28 Version: v1.0