Multi-AI Debugging

Overview

multi-ai-debugging provides systematic debugging workflows using multiple AI models as specialized agents. Based on 2024-2025 best practices for AI-assisted debugging with multi-agent architectures.

Purpose: Systematic root cause analysis and fix generation using AI ensemble

Pattern: Task-based (6 independent debugging operations)

Key Principles (validated by tri-AI research):

1. Multi-Agent Council - Specialized agents debate root causes before consensus
2. Evaluator/Critic Loops - Fix agent + critic agent verify solutions
3. Trace-Aware Analysis - Full execution context, not just error messages
4. Semantic Log Analysis - LLM understanding beyond regex matching
5. Cross-Stack Correlation - Connect frontend, backend, infra issues
6. Auto-Remediation - Self-healing patterns where safe