AI Features
RabbitQA leverages artificial intelligence to accelerate and enhance quality assurance.
AI Capabilities Overview
| Feature | Module | Description |
|---|---|---|
| Document Analysis | Analyzer | Evaluate requirement quality |
| PBI Generation | Analyzer | Auto-create product backlog items |
| Test Generation | CaseWriter | Generate test cases from requirements |
| Smart Suggestions | All | Context-aware recommendations |
Document Analysis
How It Works
- Upload - Submit requirement document
- Parse - Extract text and structure
- Analyze - AI evaluates multiple criteria
- Score - Generate quality score (0-100)
- Recommend - Suggest improvements
Analysis Criteria
| Criteria | Weight | Description |
|---|---|---|
| Clarity | 25% | Clear and unambiguous language |
| Completeness | 25% | All necessary details included |
| Testability | 20% | Can be verified through testing |
| Consistency | 15% | No contradictions |
| Structure | 15% | Well-organized content |
Quality Scores
| Score | Rating | Meaning |
|---|---|---|
| 90-100 | Excellent | Ready for implementation |
| 70-89 | Good | Minor improvements needed |
| 50-69 | Fair | Significant gaps to address |
| 0-49 | Poor | Major revision required |
PBI Generation
Generation Process
Document → AI Analysis → Extract Features → Create PBIs → Review
PBI Structure
AI-generated PBIs include:
**Title:** User Authentication via Email **Type:** Feature **Priority:** High
**Description:** As a user, I want to log in using my email and password so that I can access my
account securely.
**Acceptance Criteria:**
- [ ] Email field validates format
- [ ] Password minimum 8 characters
- [ ] Show error for invalid credentials
- [ ] Lock account after 5 failed attempts
Test Case Generation
AI Test Generator
Generate test cases from:
- Requirement documents
- User stories / PBIs
- API specifications
- Existing test cases (variations)
Generation Options
| Option | Description |
|---|---|
| Coverage Level | Basic, Standard, Comprehensive |
| Test Types | Functional, Edge Cases, Negative |
| Output Format | Steps, BDD (Gherkin), Automated |
Example Output
Input: User login requirement
Generated Tests:
Scenario: Successful login with valid credentials
Given the user is on the login page
When they enter valid email "[email protected]"
And they enter valid password "Password123"
And they click the login button
Then they should be redirected to the dashboard
Scenario: Login fails with incorrect password
Given the user is on the login page
When they enter valid email "[email protected]"
And they enter incorrect password "wrong"
And they click the login button
Then they should see error "Invalid credentials"
Smart Suggestions
Context-Aware Help
AI provides suggestions based on:
- Current action being performed
- Historical patterns
- Team practices
- Industry standards
Suggestion Types
| Type | Example |
|---|---|
| Missing Steps | "Consider adding cleanup step" |
| Similar Tests | "3 similar tests exist" |
| Coverage Gaps | "No negative tests for this feature" |
| Best Practices | "Add explicit expected result" |
AI Settings
Configuration Options
ai:
enabled: true
model: 'gpt-4'
language: 'en'
analysis:
autoAnalyze: true
minScore: 70
generation:
defaultCoverage: 'standard'
includeEdgeCases: true
maxTestsPerRequirement: 10
Privacy & Data
- Documents processed in isolated environments
- No training on customer data
- Data deleted after processing
- Opt-out available for AI features
bilgi
AI features can be disabled at organization or project level in Settings.
