skills.rest
woink
woinkCOMMUNITY
Explore all skills by this author →
GitHub

harness-evaluator

Evaluate running applications against sprint contracts and write structured QA reports.

Updated Aug 23, 2026
Search Tags:#testing#qa#code-quality#playwright#harness#regression-testing#polish-mode
One-click install
npx skills add https://github.com/woink/harness-skills-plugin --skill harness-evaluator
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: harness-evaluator
Source: https://github.com/woink/harness-skills-plugin/tree/main/skills/harness-evaluator
Command: npx skills add https://github.com/woink/harness-skills-plugin --skill harness-evaluator

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

The harness-evaluator eliminates biased self-review by independently testing running applications against explicit sprint contracts and calibrated grading criteria, ensuring build quality, regression safety, and measurable readiness for release.

Core Features & Use Cases

  • Contract Review & Testability Checks: Validates sprint contracts for completeness, measurability, and start commands before coding begins and produces a structured contract-review artifact.
  • Sprint QA & Regression Testing: Runs live-app checks, automated test suites, and regression checks against CODEBASE_AUDIT.md for existing projects, producing a detailed .harness/QA_REPORT.md.
  • Polish Mode & Calibration: Performs iterative polish evaluations with sub-criteria, drift detection against POLISH_HISTORY.md, and pivot suggestions when stalled; consults embedded calibration anchors and project-specific overrides.
  • Integration Requirements: Uses Playwright MCP for web evaluation, consults the frontend-design skill for design grading, and requires Context7 for authoritative API documentation checks.

Quick Start

Run harness-evaluator on the current sprint to produce a calibrated .harness/QA_REPORT.md (or POLISH_REPORT.md in polish mode) and a PASS/FAIL verdict with actionable remediation.

Frequently Asked Questions about harness-evaluator

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

FAQPage Schema
How do I run regression testing for running web apps against sprint contracts?▼

You can run regression testing against sprint contracts by evaluating live web applications using Playwright MCP to execute live-app checks, automated test suites, and regression checks against CODEBASE_AUDIT.md, producing a detailed QA_REPORT.md with calibrated scores.

What is automated QA evaluation and how does it eliminate biased self-review?▼

Automated QA evaluation independently tests running applications against explicit sprint contracts and calibrated grading criteria. By using an external harness evaluator instead of self-review, it ensures objective build quality, regression safety, and measurable release readiness.

Does the harness-evaluator work with Playwright MCP for web evaluation?▼

Yes, the harness-evaluator uses Playwright MCP for web interactions to evaluate running applications. It also consults the frontend-design skill for design grading and Context7 for authoritative API documentation checks during the evaluation process.

How do I validate sprint contracts for completeness and testability before coding?▼

You validate sprint contracts by running a contract review that checks for completeness, measurability, and start commands before coding begins. This process produces a structured contract-review artifact to ensure the sprint requirements are testable.

What is polish mode for iterative QA evaluations and drift detection?▼

Polish mode performs iterative QA evaluations with sub-criteria and drift detection against POLISH_HISTORY.md. It consults embedded calibration anchors and project-specific overrides, offering pivot suggestions when progress stalls, and writes a POLISH_REPORT.md.

Can I use harness-evaluator for existing projects or only new sprint builds?▼

You can use harness-evaluator for both new sprint builds and existing projects. For existing projects, it performs regression checks against CODEBASE_AUDIT.md, while new sprints are evaluated against their specific sprint contracts to produce a PASS/FAIL verdict.

Related Skills

Explore More Software Engineering Skills→
labs21-devlabs21-dev
28

design-proposer

Draft a bounded sprint contract from design.md and expose an awaiting_human approval gate.

Official
Advanced
h-4vokh-4vok

repo-validate

Run lint, typecheck, test, smoke, and build checks across a repository and emit a pass/fail report.

Community
Intermediate
TeiNamTeiNam
12

documentation-lookup

Fetch live library and framework documentation via Context7 MCP.

Community
Advanced
VitorAndTxrVitorAndTxr

qa

Guide structured testing across planning, design, review, and execution phases.

Community
Advanced