harness-evaluator

Audit post-implementation stories with ordered checks and a go/no-go verdict.

Updated Apr 18, 2026
One-click install
npx skills add https://github.com/ZhaoyuWu/Website-Wzy --skill harness-evaluator-zhaoyuwu
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: harness-evaluator
Source: https://github.com/ZhaoyuWu/Website-Wzy/tree/main/codex-skills/harness-evaluator
Command: npx skills add https://github.com/ZhaoyuWu/Website-Wzy --skill harness-evaluator-zhaoyuwu

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill standardizes the post-implementation audit process by enforcing a strict evaluator role, ensuring the task is read first, and delivering a clear go/no-go decision.

Core Features & Use Cases

  • Role-based auditing: you assume the auditor branch owner role to guide the review.
  • Ordered audit steps: security, functionality and tests, performance, and principles.
  • Findings classification: categorize issues as P0, P1, or P2 and decide Go/No-Go.
  • Context-driven decisions: outputs reflect task context and evaluation rules, suitable for /evaluator workflows.

Quick Start

Provide the task details and run the evaluator to read the task, perform the ordered audit, and return a go/no-go decision.

Frequently Asked Questions about harness-evaluator

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

FAQPage Schema
How do I audit a software workflow and decide go or no-go after implementation?

To audit a software workflow and decide go or no-go, provide your task details to an evaluator that reads the context, checks security, functionality, tests, performance, and principles, then classifies findings as P0, P1, or P2 to return a clear decision.

What is a P0 P1 P2 severity classification in post-implementation auditing?

P0, P1, and P2 severity classification in post-implementation auditing categorizes discovered issues by criticality to determine if a workflow passes or fails, directly driving the final go/no-go verdict based on the highest severity finding.

What steps should I follow to evaluate task functionality and security before release?

Evaluating task functionality and security before release requires following ordered audit steps: run security checks first, verify functionality and tests next, assess performance, and review principles to classify issues and output a go/no-go decision.

Can I use a structured evaluator workflow for software-quality checks without manual review?

Yes, you can use a structured evaluator workflow for software-quality checks without manual review by assigning the auditor role to an automated evaluator that ingests task context and applies disciplined sequential checks to output a verdict.

When do I need a go-no-go decision process for my software development workflow?

You need a go-no-go decision process for your software development workflow when you require standardized post-implementation auditing to enforce strict quality gates, ensuring security, functionality, performance, and principles are verified before release.