skills.rest
masanao-ohba
masanao-ohbaCOMMUNITY
Explore all skills by this author →
GitHub

completion-evaluator

Automates evaluation of deliverables against acceptance criteria to determine pass/fail verdicts.

2|Updated Oct 19, 2025
Search Tags:#automation#QA#workflows#evidence-based#acceptance-criteria#evaluate#verdict
One-click install
npx skills add https://github.com/masanao-ohba/claude-manifests --skill completion-evaluator
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: completion-evaluator
Source: https://github.com/masanao-ohba/claude-manifests/tree/main/skills/generic/completion-evaluator
Command: npx skills add https://github.com/masanao-ohba/claude-manifests --skill completion-evaluator

SYSTEM DOCUMENTATION & REQUIREMENTS

## What problem does it solve? The Completion Evaluator solves the problem of subjective or inconsistent verdicts when validating whether a deliverable meets defined acceptance criteria. It provides an objective, evidence-based framework to determine PASS/FAIL, justify decisions, and offer concrete improvement guidance.

## Core Features & Use Cases

  • Objective evaluation of deliverables using completeness, quality, and functionality indicators.
  • Evidence collection and verdict justification.
  • Use Case: When a sprint artifact is submitted, the evaluator checks criteria and outputs a structured evaluation report with next steps for rework or approval.

### Quick Start

  • Provide the deliverable artifact and acceptance criteria to the evaluator.
  • Review the generated evaluation document containing overall verdict, indicator results, and feedback.

Frequently Asked Questions about completion-evaluator

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

FAQPage Schema
How do I automate deliverable evaluation against acceptance criteria?▼

You can automate deliverable evaluation by providing the artifact and acceptance criteria to an evidence-based framework. The evaluator applies completeness, quality, and functionality indicators to generate an objective PASS/FAIL verdict with actionable feedback.

What is evidence-based evaluation for software QA workflows?▼

Evidence-based evaluation is an objective framework that collects indicators and applies thresholds to determine pass/fail verdicts. It justifies decisions with collected evidence, replacing subjective validation with structured assessments for software project artifacts.

How do I evaluate sprint artifacts for approval or rework?▼

To evaluate sprint artifacts, submit the deliverable and its acceptance criteria. The evaluator outputs a structured report containing the overall verdict, individual indicator results, and concrete next steps for either approval or rework.

Can I use this evaluation framework for code, documentation, and product artifacts?▼

Yes, the evaluation framework applies to code, documentation, and product artifacts across software projects. It assesses completeness, quality, and functionality indicators to provide objective assessments regardless of the deliverable type.

How does the evaluator integrate with a workflow orchestrator?▼

The evaluator integrates with a workflow orchestrator by outputting a structured evaluation format. This structured output contains the overall verdict, indicator results, and feedback, allowing downstream automation to trigger approvals or rework actions.

What is the best way to remove subjective verdicts from acceptance criteria validation?▼

The best way to remove subjective verdicts is applying an evidence-based evaluation framework with defined indicators and thresholds. This replaces inconsistent manual validation with objective PASS/FAIL decisions and concrete improvement guidance.

Related Skills

Explore More Software Engineering Skills→
jshsakurajshsakura
22

business-analyst

Clarify requirements, normalize scope, and define acceptance criteria from messy inputs.

Community
Intermediate
Yeachan-HeoYeachan-Heo
32.6k

review

Evaluate plan files in .omx/plans/ with Critic-based review verdicts.

Community
Basic
liuzeming-yuxiliuzeming-yuxi
9

superclaw-verify

Cross-checks Claude Code outputs against spec and plan via automated tests and validations.

Community
Advanced
Busboom-FooBusboom-Foo

code-review

Assess code changes against acceptance criteria with a two-phase review workflow.

Official
Advanced