solution-evaluation

Evaluate design documents against a five-dimension rubric and output badcase lists.

52|3|Updated Feb 14, 2026
One-click install
npx skills add https://github.com/ZTE-AICloud/Co-OmniSpec --skill solution-evaluation
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: solution-evaluation
Source: https://github.com/ZTE-AICloud/Co-OmniSpec/tree/main/skills/solution-evaluation
Command: npx skills add https://github.com/ZTE-AICloud/Co-OmniSpec --skill solution-evaluation

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a five-dimension quality evaluation rubric for design documents, aligned to the latest seven-chapter structure and content richness standards. It is used by the design step to evaluate design.md outputs, generating badcase lists and surface_reason, and explicitly does not produce root-cause analysis or prompt-diagnosis guidance.

Core Features & Use Cases

  • Five-dimension scoring across structure, semantic accuracy, business_rules, compliance, and overall quality.
  • Checks alignment with the spec, coverage of key concepts, and consistency of design decisions.
  • Produces a formal evaluation report with bad cases and surface reasons to guide improvements.

Quick Start

Follow the instruction to integrate the evaluation rubric into your design-review workflow.

Frequently Asked Questions about solution-evaluation

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

FAQPage Schema
How do I evaluate a design document against a structured quality rubric?

To evaluate a design document against a quality rubric, apply a five-dimension scoring framework covering structure, semantic accuracy, business rules, compliance, and overall quality. This generates a formal assessment with a badcase list and surface reasons to guide improvements.

What is the best way to identify gaps in a design proposal during a phase review?

Identifying gaps in a design proposal during a phase review requires checking alignment with the spec, coverage of key concepts, and consistency of design decisions. This evaluation outputs a formal report detailing bad cases and surface reasons for improvement opportunities.

Can I use this evaluation to perform root-cause analysis on bad design decisions?

No, this evaluation cannot perform root-cause analysis on bad design decisions. It explicitly produces a badcase list with surface reasons to identify gaps and improvement opportunities, but does not provide root-cause analysis or prompt-diagnosis guidance.

How does a multi-dimension rubric assess semantic accuracy and compliance in architecture notes?

A multi-dimension rubric assesses semantic accuracy and compliance in architecture notes by enforcing scoring across five specific dimensions. It checks alignment with the spec, coverage of key concepts, and consistency of design decisions to produce a formal quality assessment.

When do I need a formal quality assessment for design.md outputs?

You need a formal quality assessment for design.md outputs during design-phase reviews of design proposals and architecture notes. Applying a five-dimension rubric ensures your documentation meets structure, semantic accuracy, business rules, compliance, and overall quality standards.

Does the design evaluation skill require external dependencies or components to run?

No, the design evaluation skill requires no external dependencies or components to run. You can integrate the evaluation rubric directly into your design review workflow to assess design documents and generate badcase lists without additional environment setup.