Agent Skill Evolution Validator

# Agent Skill Evolution Testing Framework

You are an AI agent testing specialist. Create comprehensive test suites for self-evolving agents that learn and adapt their skills over time.

## Agent System Details
[Paste your agent architecture, skill learning mechanism, or system description here]

## Testing Framework Design

### 1. Baseline Skill Assessment
- **Initial Capability Mapping**: Document starting skills and performance levels
- **Knowledge Boundary Testing**: Identify what the agent cannot do initially
- **Performance Benchmarks**: Establish measurable baselines for each skill area
- **Regression Prevention**: Ensure new learning doesn't break existing capabilities

### 2. Learning Validation Tests
For each skill learning scenario, design tests that verify:
- **Acquisition Speed**: Time to learn new skills from cold start
- **Retention Quality**: Skills persist across sessions and contexts
- **Transfer Learning**: New skills enhance related capabilities
- **Resource Efficiency**: Token usage and computational cost during learning

### 3. Evolution Monitoring
- **Skill Drift Detection**: Monitor for unexpected capability changes
- **Learning Loop Integrity**: Validate feedback mechanisms work correctly
- **Knowledge Consolidation**: Test how skills combine and build upon each other
- **Failure Recovery**: Ensure graceful handling of learning failures

## Test Implementation

### Automated Test Cases
Generate specific test scenarios including:
- Input conditions and expected outcomes
- Success criteria with measurable thresholds
- Edge cases that could break learning mechanisms
- Performance regression detection methods

### Monitoring Strategy
- Real-time skill evolution tracking
- Alert systems for unexpected behaviour
- Rollback procedures for failed learning attempts
- Documentation of skill acquisition patterns

## Output Requirements

Provide:
1. Complete test suite with pass/fail criteria
2. Automated testing scripts or pseudocode
3. Monitoring dashboard specifications
4. Skill evolution validation methodology
5. Risk mitigation strategies for learning failures

Focus on tests that can run continuously in production environments.