skills.rest
EricAnalysis
EricAnalysisCOMMUNITY
Explore all skills by this author →
GitHub

eightforge-code-reviewer

Reviews EightForge pull requests and designs for architecture, authority, and truth-provenance violations.

Updated Mar 8, 2026
Search Tags:#architecture#code review#pull request#validation#audit#provenance#eightforge
One-click install
npx skills add https://github.com/EricAnalysis/eightforge-os --skill eightforge-code-reviewer-ericanalysis
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: eightforge-code-reviewer
Source: https://github.com/EricAnalysis/eightforge-os/tree/main/.cursor/skills/eightforge-code-reviewer
Command: npx skills add https://github.com/EricAnalysis/eightforge-os --skill eightforge-code-reviewer-ericanalysis

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Reviewing broad pull requests and architecture changes in the EightForge platform risks missing cross-cutting violations such as parallel authority paths, fabricated source truth, shadow-code leakage, and broken layer boundaries. This Skill applies a consistent umbrella review doctrine so every change is checked against the same architectural invariants. ## Core Features & Use Cases - Authority Classification: Classifies every reviewed path and finding as PRODUCTION AUTHORITY, PRODUCTION COMPATIBILITY, SHADOW IMPLEMENTED, DESIGN TARGET, or SUPERSEDED/REJECTED. - Architecture Checklist: Enforces one-way Extraction-to-Audit flow, truth-record separation, shadow isolation, compatibility containment, and minimal-diff scope. - Structured Verdict Output: Produces a fixed report format with verdict, severity-ordered key issues, minimal fixes, regression risks, suggested tests, and one positive note. - Use Case: When reviewing a full PR that touches validation, document pipelines, and project-facing projections, use this Skill to trace writers, consumers, and fallbacks, then route domain-specific checks to the specialized EightForge reviewers. ## Quick Start Review this pull request for EightForge architecture violations and produce a verdict with severity-ordered issues and minimal fixes.

Frequently Asked Questions about eightforge-code-reviewer

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

FAQPage Schema
How do I review a large pull request for architecture violations?▼

Use this reviewer to classify every affected path by authority class, trace writers and consumers, and check the one-way Extraction-to-Audit flow. It outputs a verdict with severity-ordered issues, minimal fixes, and regression risks.

What is the difference between production authority and production compatibility in code review?▼

Production authority is the canonical persisted or computed truth path, while production compatibility covers legacy projections current readers still consume. Compatibility paths must be contained, never widened or presented as the target architecture.

When should I use the umbrella reviewer versus the specialized EightForge reviewers?▼

Use the umbrella reviewer for full PRs, broad designs, or changes spanning multiple domains. Route domain-specific checks like migrations, audit storage, RLS, performance, or UX mechanics to the corresponding specialized reviewer.

Can shadow code influence live validation or decisions?▼

No. Shadow code and artifacts must remain lifecycle-isolated and cannot influence live facts, validation, decisions, execution, or UI. Promotion to production requires an explicit recorded cutover decision and its gates.

What output format does the architecture review produce?▼

The review produces a fixed structure: a verdict of Pass, Pass with Concerns, or Fail, followed by severity-ordered key issues with evidence, minimal fixes, regression risks, suggested tests, and exactly one positive note.

Related Skills

Explore More Software Engineering Skills→
macollins27macollins27

plan

Produces design.md, plan.md and codebase-context.md from a spec and codebase context.

Community
Advanced
ZaxbyHubZaxbyHub

council-advisory-triage

Classifies council review findings as blocking or advisory to prevent unnecessary rework.

Official
Intermediate
nguyenphpnguyenphp
41

parallel-agents

Coordinate specialized AI agents for complex task execution in Claude Code.

Community
Advanced
LedgerHQLedgerHQ
615

ddd-data-layer-advanced

Structures Ledger Wallet data layers where one API response serves multiple domain entities.

Official
Intermediate