How does the scoring system work?

It uses a weighted rubric across five core dimensions: Accuracy (25%), Completeness (20%), Usefulness (25%), Clarity (15%), and Freshness (15%), resulting in a final score from 1.0 to 5.0.

What outputs can I evaluate with this skill?

You can evaluate almost any professional output, including strategic plans, technical documentation, research summaries, pitches, and AI-generated code explanations.

Does it provide specific fix suggestions?

Yes, every evaluation produces a ranked list of improvements categorized by priority, explaining exactly why an issue matters and how to fix it.

Why use a rubric instead of a general review?

A rubric ensures consistency, prevents grade inflation, and requires specific evidence for every claim, making the feedback objective and actionable.

Can I use this for technical code documentation?

Absolutely. It includes specific extensions to assess technical docs for reproducibility, command accuracy, and coverage of edge cases.

Strategic Output Evaluation

Name: Strategic Output Evaluation
Author: nsalvacao

bynsalvacao

0•

콘텐츠 관리

Evaluates and scores documents, plans, and code outputs using a structured, rubric-based framework to ensure production-grade quality.

The Strategic Output Evaluation skill provides a systematic approach to quality assurance for any AI-generated or human-authored content. By applying a weighted rubric across dimensions like accuracy, usefulness, and clarity, it generates objective scores and prioritized improvement lists. This tool is essential for professionals who need to audit strategic plans, technical documentation, or research summaries before they are shared externally, ensuring every output meets high standards of reliability and professionalism.

주요 기능

010 GitHub stars

02Evidence-based reporting requiring specific quotes for every score

03Weighted 5-dimension rubric for objective quality scoring

04Automated quality classification and sharing recommendations (A-F grades)

05Specific evaluation extensions for strategy, tech docs, and research

06Prioritized improvement lists categorizing 'Must Fix' vs 'Nice to Have'

사용 사례

01Quality-checking AI-generated research reports for bias and source credibility

02Reviewing a strategic business plan before a high-stakes board presentation

03Auditing technical documentation for accuracy, reproducibility, and edge cases

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add nsalvacao/nsalvacao-claude-code-plugins evaluation

For use in Claude.ai and ChatGPT

주요 기능

010 GitHub stars

02Evidence-based reporting requiring specific quotes for every score

03Weighted 5-dimension rubric for objective quality scoring

04Automated quality classification and sharing recommendations (A-F grades)

05Specific evaluation extensions for strategy, tech docs, and research

06Prioritized improvement lists categorizing 'Must Fix' vs 'Nice to Have'

사용 사례

01Quality-checking AI-generated research reports for bias and source credibility

02Reviewing a strategic business plan before a high-stakes board presentation

03Auditing technical documentation for accuracy, reproducibility, and edge cases

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add nsalvacao/nsalvacao-claude-code-plugins evaluation

For use in Claude.ai and ChatGPT