Skip to content

EXAMPLES.md

Latest commit

 

History

History
157 lines (110 loc) · 3.99 KB

EXAMPLES.md

File metadata and controls

157 lines (110 loc) · 3.99 KB
Raw

Example Data

This project includes example data to help you understand how the system works.

Example Files Included

.dev/ - Learning Data

error_database.json

  • 5 example errors across 3 apps
  • 4 identified patterns (JSON syntax, security, missing files, tests)
  • Shows how errors are tracked and patterns detected

skill_history.json

  • Empty initially, ready for your first skill generation
  • Tracks all skill versions and their effectiveness

AUTO_UPDATES.md

  • Current analysis of patterns
  • Recommendations for skill generation
  • Next steps in the learning cycle

results/ - Test Results

3 example test results:

  • EXAMPLE_APP_001_result.json - Grade C (75/100) - Has JSON errors, missing README, no tests
  • EXAMPLE_APP_002_result.json - Grade D (65/100) - Security issues, low test coverage
  • EXAMPLE_APP_003_result.json - Grade A (95/100) - Excellent quality, comprehensive tests

test-criteria/ - Validation Criteria

EXAMPLE_APP_001-criteria.json

  • Sample criteria for a web application
  • Shows required files and validation requirements

use-cases/ - Common Patterns

example-web-app.json

  • Basic web application use case
  • Requirements and validation criteria

example-api-service.json

  • REST API service use case
  • More comprehensive requirements

How to Use These Examples

1. Explore the Data

# View error patterns
cat .dev/comparison/error_database.json

# View test results
cat results/EXAMPLE_APP_001_result.json

2. Generate Your First Skill

The example data has enough patterns to generate a skill:

# Analyze patterns
python3 skill_updater.py analyze

# Generate skill
python3 skill_updater.py generate

This will create skills/app-quality-json-security-tests/SKILL.md based on the example patterns.

3. Test with Your Own Apps

Replace the examples with your own applications:

# Setup your app
python3 setup_test.py MY_APP

# Copy your app
cp -r /path/to/your/app test-apps/MY_APP/

# Test it
python3 automate_test.py --evaluate test-apps/MY_APP --app-id MY_APP

4. Learn from Real Data

As you test more apps:

  • Errors accumulate in error_database.json
  • Patterns emerge automatically
  • Skills improve based on real failures
  • Results show progress over time

Example Workflow

# 1. See what patterns exist
python3 skill_updater.py analyze

# 2. Generate skill from examples
python3 skill_updater.py generate

# 3. View the generated skill
cat skills/app-quality-*/SKILL.md

# 4. Test your own app
python3 automate_test.py --evaluate test-apps/your-app --app-id YOUR_APP

# 5. Measure improvement
python3 skill_updater.py measure --version v1

Cleaning Up Examples

To start fresh with your own data:

# Remove example data (keeps structure)
rm -f .dev/comparison/error_database.json
rm -f .dev/learning/skill_history.json
rm -f results/*.json
rm -f test-criteria/*.json
rm -f use-cases/*.json

# Recreate empty files
echo '{"errors": [], "patterns": {}}' > .dev/comparison/error_database.json
echo '{"updates": [], "versions": {}, "effectiveness": {}}' > .dev/learning/skill_history.json

What Each Example Shows

EXAMPLE_APP_001 (Grade C)

  • Problem: JSON errors, missing documentation, no tests
  • Learning: Shows common beginner mistakes
  • Skill Opportunity: JSON validation, file completeness

EXAMPLE_APP_002 (Grade D)

  • Problem: Hardcoded secrets, low test coverage
  • Learning: Security issues are critical
  • Skill Opportunity: Secure coding practices, testing

EXAMPLE_APP_003 (Grade A)

  • Problem: None - this is the target quality
  • Learning: What a well-built app looks like
  • Skill Opportunity: Use as a reference for best practices

Next Steps

  1. Explore: Look through the example files
  2. Generate: Create your first skill from examples
  3. Test: Add your own applications
  4. Learn: Watch the system improve over time

Tip: Keep the examples around as reference, or delete them once you understand the system!