model-tracking-protocol

What problem does it solve?

This Skill ensures that when multiple AI models are run in parallel for validation, their execution is meticulously tracked from pre-launch to results presentation, preventing incomplete reviews and lost data.

Core Features & Use Cases

• Mandatory Pre-Launch Checklist: Enforces critical setup steps before model execution.
• Structured Tracking Tables: Provides templates for logging model status, timing, issues, and quality.
• Failure Documentation: Standardizes the reporting of model failures with error messages and retry status.
• Consensus Analysis: Compares findings across models to identify agreement and divergence.
• Results Presentation: Ensures all necessary components (summary, performance, issues, details) are included in the final output.
• Use Case: When evaluating three different AI models for code review, this Skill ensures you create a tracking table, record when each model starts and finishes, document any errors, and finally compare their findings to reach a consensus on the most critical issues.

Quick Start

Execute the pre-launch script to set up session tracking before running any models.