harness-summary

Name: harness-summary
Availability: InStock
Author: ats-kinoshita-iso

Analyze multi-sprint evaluation data to produce pass rate and failure pattern insights.

Updated Apr 11, 2026

Search Tags:#performance #metrics #evaluation #assessment #trend analysis #sprint review

Ready to run

npx skills add https://github.com/ats-kinoshita-iso/trine-eval --skill harness-summary

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: harness-summary
Source: https://github.com/ats-kinoshita-iso/trine-eval/tree/main/skills/harness-summary
Command: npx skills add https://github.com/ats-kinoshita-iso/trine-eval --skill harness-summary

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill addresses the challenge of synthesizing evaluation results from multiple sprints to provide comprehensive insights into system performance and trends.

Core Features & Use Cases

Evaluation Metric Aggregation: Calculates overall pass rates, pass@k, pass^k, and detects anomalies across sprints.
Trend & Pattern Analysis: Identifies performance improvements or regressions over time and recurring failure patterns.
Use Case: A project manager reviews the evaluation history to determine if the system's reliability is improving and identifies bottlenecks needing attention.

Quick Start

Generate a comprehensive evaluation summary for the current sprint history to guide next development cycles.

Frequently Asked Questions about harness-summary

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

FAQPage Schema

How do I analyze multi-sprint evaluation data to identify performance trends?▼

Multi-sprint evaluation data analysis aggregates pass rates, pass@k, and pass^k metrics to identify performance trends and recurring failure patterns for continuous system improvement. It processes evaluation reports, configs, and contracts to detect anomalies across sprints.

What is cross-sprint evaluation metric aggregation for AI workflows?▼

Cross-sprint evaluation metric aggregation calculates overall pass rates and detects anomalies across multiple evaluation cycles in AI workflows. It synthesizes historical evaluation reports to provide comprehensive insights into system reliability and consistency.

How do I generate a sprint review summary from evaluation reports and contracts?▼

Generate a sprint review summary by analyzing evaluation reports, configs, and contracts to calculate pass rates and detect failure patterns. This produces insights on consistency and regressions to guide next development cycles.

Can I detect recurring failure patterns and regressions from project retrospectives?▼

Yes, recurring failure patterns and regressions are detected by analyzing multi-sprint evaluation data during project retrospectives. The analysis identifies performance improvements or degradations over time and highlights bottlenecks needing attention.

Do I need evaluation configs and contracts to track system stability improvements?▼

Yes, evaluation configs and contracts are required to track system stability improvements accurately. They provide the necessary baseline and criteria to synthesize multi-sprint results and determine if system reliability is improving.