eval-projects

Analyze Ralph loop logs and reports to suggest critical and nice-to-have improvements.

3|Updated Feb 14, 2026
One-click install
npx skills add https://github.com/mattwoodco/skills --skill eval-projects
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: eval-projects
Source: https://github.com/mattwoodco/skills/tree/main/skills/eval-projects
Command: npx skills add https://github.com/mattwoodco/skills --skill eval-projects

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill addresses the challenge of understanding and improving the performance and efficiency of AI-driven project loops by analyzing their execution logs and progress reports.

Core Features & Use Cases

  • Comprehensive Analysis: Reviews loop structure, skill combinations, outcomes, costs, and logs.
  • Actionable Recommendations: Provides clear suggestions for critical and nice-to-have improvements.
  • Use Case: After running multiple AI projects, use this Skill to identify which steps are consistently failing, which skill combinations are inefficient, and where costs are escalating, then receive concrete steps to optimize future runs.

Quick Start

Use the eval-projects skill to review the evaluation tracking from all ralph loops and their logs.

Frequently Asked Questions about eval-projects

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I evaluate AI loop performance using execution logs?

Evaluating AI loop performance requires analyzing execution logs, GENERATED.md, and PROGRESS.md files to identify inefficient skill combinations and operational failures. This skill processes loop catalogs and report history to pinpoint issues in loop structure and tooling.

What is the best way to debug failing AI project loops?

Debugging failing AI project loops involves reviewing loop catalogs and associated logs to detect structural issues and escalating costs. This skill analyzes evaluation tracking to suggest critical and nice-to-have improvements for future runs.

Can I analyze multiple AI project loops at once to find inefficient skill combinations?

Analyzing multiple AI project loops is supported by processing evaluation tracking across all project directories. It assesses loop structure, skill combinations, and outcomes to identify which steps consistently fail and where performance tuning is needed.

Do I need specific log and report files to assess AI project performance?

Assessing AI project performance requires access to project loop directories and associated log and report files. The skill needs these inputs, including GENERATED.md and PROGRESS.md, to provide a comprehensive evaluation of operations and costs.

Why does my AI project loop have escalating costs and inefficient operations?

Escalating costs and inefficient operations in AI loops stem from suboptimal skill combinations and tooling issues. This skill analyzes report history and logs to identify these exact cost drivers and suggests concrete steps to optimize future runs.

What limitations exist when using eval-projects for loop analysis?

Loop analysis is limited by its dependency on accessible project loop directories and associated log files. Without comprehensive access to GENERATED.md, PROGRESS.md, and report history, the skill cannot accurately assess operations or suggest improvements.