test-evaluation

Evaluate and debug test suites for flakiness and regression coverage.

4|1|Updated Mar 11, 2026
One-click install
npx skills add https://github.com/jcdendrite/claude-config --skill test-evaluation
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: test-evaluation
Source: https://github.com/jcdendrite/claude-config/tree/main/claude/.claude/skills/test-evaluation
Command: npx skills add https://github.com/jcdendrite/claude-config --skill test-evaluation

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you evaluate, debug, and improve existing test suites that have become unreliable, slow, or insufficient at catching regressions.

Core Features & Use Cases

  • Test pyramid assessment: Identifies when tests are stuck in integration/E2E layers and provides guidance for moving them down toward faster, more deterministic unit tests.
  • Flaky test diagnosis: Helps pinpoint root causes such as isolation failures, external service timing issues, locale/timezone sensitivity, async synchronization mistakes, and nondeterministic ordering.
  • Actionable refactoring guidance: Recommends safer remediation tactics like layer-correcting tests, quarantining when needed, and avoiding destructive changes without replacement coverage.

Use Case: Your CI intermittently fails while local runs pass; use the Skill to determine which tests are likely flaky for systemic reasons (e.g., network timeouts, shared state, or nondeterministic data), then plan the most appropriate fix at the correct test layer.

Quick Start

Ask the AI to evaluate and debug your existing test suite by identifying weak coverage areas and diagnosing the most likely causes of flakiness in the failing tests.

Frequently Asked Questions about test-evaluation

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

FAQPage Schema
How do I diagnose flaky tests causing intermittent CI failures?

Diagnosing flaky tests involves checking root causes like isolation failures, external service timing, async synchronization mistakes, and nondeterministic data ordering. The Skill evaluates these scenarios to pinpoint why CI intermittently fails while local runs pass.

What is the test pyramid and how do I fix tests stuck in integration layers?

The test pyramid prioritizes faster, deterministic unit tests over slow integration or E2E tests. The Skill assesses your suite, identifies tests stuck in higher layers, and provides guidance for moving them down toward more reliable unit coverage.

How do I debug slow test suites and improve regression coverage?

Debugging slow test suites requires evaluating existing coverage against the intended test pyramid to find weak areas. The Skill evaluates suite quality, identifies slow or network-dependent tests, and recommends correct-layer refactors to improve coverage.

Should I delete flaky tests or is there a safer way to handle them?

Deleting flaky tests blindly removes regression coverage. The Skill uses guardrails to quarantine flaky coverage when needed and recommends replacing it with safer, layer-correct tests rather than making destructive changes without replacements.

What causes nondeterministic test failures related to locale or timezone?

Nondeterministic test failures often stem from locale or timezone sensitivity, shared state, or network timeouts. The Skill diagnoses these specific root causes in failing tests and recommends the most appropriate fix at the correct test layer.

Can this Skill evaluate test-code quality for my existing test suite?

Yes, the Skill performs test-code reviews of suite quality versus the intended test pyramid. It evaluates existing test suites to resolve uncertainty about regression coverage and identifies weak coverage areas requiring refactoring.