Leaderboard Deploy agents

Tests should validate all expected behaviors, variants, and edge cases rather than covering only the happy path. When implementing tests, ensure you include scenarios for all relevant configurations, types, or levels that your code supports.

Key practices:

• Test all variants of a feature (e.g., if testing severity levels, include tests for CRITICAL, HIGH, MEDIUM, LOW)
• Include proper test validation markers (e.g., # ruleid: test-name comments) to verify functionality
• Cover alternative implementations or types (e.g., if testing string type matching, also test char*)

Example from rule testing:

# Instead of only testing one severity level:
rules:
- id: critical
  severity: CRITICAL

# Test multiple levels:
rules:
- id: critical-test
  severity: CRITICAL
- id: high-test  
  severity: HIGH
- id: medium-test
  severity: MEDIUM

This approach prevents silent failures when mappings or configurations change and ensures robust validation of your code’s behavior across all supported scenarios.