skills.rest
hungv47
hungv47COMMUNITY
Explore all skills by this author →
Website𝕏@hungv47GitHub

review-work

Review completed changes with layered evaluation and generate dated artifact reports.

14|Updated Apr 1, 2026
Search Tags:#dashboard#quality#review#artifact#log#fresh-eyes
One-click install
npx skills add https://github.com/hungv47/meta-skills --skill review-work-hungv47
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: review-work
Source: https://github.com/hungv47/meta-skills/tree/main/skills/meta/review-work
Command: npx skills add https://github.com/hungv47/meta-skills --skill review-work-hungv47

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Independent, objective post-implementation reviews that uncover issues missed during development and surface actionable improvements.

Core Features & Use Cases

  • Structured, evidence-based review prompts that separate reviewer output from implementation reasoning.
  • Layered evaluation (Layer 1 real-vs-fake verification; Layer 2 actionable triage) with a formal verdict.
  • Generates dated, immutable artifact reports and logs quality-dashboard interactions when overrides occur.

Quick Start

Perform a completed change review to trigger Fresh-Eyes; you will receive a structured verdict and recommended follow-ups.

Frequently Asked Questions about review-work

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

FAQPage Schema
What is a fresh-eyes code review and when should I use it?▼

A fresh-eyes review is an independent post-implementation evaluation that separates reviewer output from implementation reasoning to uncover missed issues. Use it after completing a change to verify code quality through structured, evidence-based prompts.

How do I perform a post-implementation review that generates an immutable report?▼

Trigger a fresh-eyes review on a completed change to execute a two-layer evaluation: real-vs-fake verification followed by actionable triage. This produces a dated, immutable artifact report at docs/forsvn/artifacts/meta/records/ with a formal verdict and follow-ups.

Can I route code quality reviews by artifact type and risk level?▼

Yes, the fresh-eyes review process routes evaluations by artifact type and risk level. It uses a generalist reviewer and an optional resolver to produce a structured verdict, automatically logging quality dashboard interactions when overrides occur.

What is the best way to log code quality dashboard overrides during a review?▼

The best way to log quality dashboard overrides is through the fresh-eyes review process, which automatically logs interactions when overrides occur during the two-layer triage, ensuring all deviations are recorded alongside the generated structured verdict.

Does the fresh-eyes review work without external dependencies?▼

Yes, the fresh-eyes review operates without external dependencies, using internal scripts and references to perform two-layer verification and triage, generating immutable artifact reports and logging dashboard overrides independently within your environment.

Related Skills

Explore More Software Engineering Skills→
grp06grp06
83

review-recent-work

Reviews the latest ExecPlan after implementation, identifies bugs and assesses usefulness.

Community
Advanced
CharlieGreenmanCharlieGreenman
1

gpd-compare-results

Compare internal results against baselines to produce decisive verdicts.

Community
Advanced
tmalcolm-0607tmalcolm-0607

council-verdict

Override Council verdicts on governance threads with mandatory rationale and consent gates.

Community
Advanced
Chang-Shih-YungChang-Shih-Yung

health

Aggregate type checker, linter, and test results into a 0-10 composite score.

Community
Advanced