Multi-AI Testing

Overview

multi-ai-testing provides test-driven development workflows with independent verification to prevent test gaming and ensure comprehensive test coverage.

Purpose: Generate high-quality tests through TDD, achieve ≥80% coverage (gate) / ≥95% (target), prevent agents from gaming their own tests

Pattern: Workflow-based (4 core workflows)

Key Innovation: Independent verification through separate test/implementation agents prevents overfitting and test gaming

Core Principles (validated by tri-AI research):

1. Test-First Development - Write tests BEFORE implementation
2. Independent Verification - Separate agents for testing vs. implementation
3. Comprehensive Coverage - ≥80% gate, ≥95% achievable with AI edge case discovery
4. Non-Deterministic Evaluation - Scoring systems, not binary pass/fail
5. Self-Healing Tests - Tests adapt to code changes (80% maintenance reduction)