skills.rest
nnsi
nnsiCOMMUNITY
Explore all skills by this author →
GitHub

review-cycle

Coordinate parallel code reviews from architecture and implementation perspectives until both approve.

1|Updated Aug 31, 2024
Search Tags:#quality-assurance#code-review#workflow-orchestration#ai-assisted#parallel-review#lgtm
One-click install
npx skills add https://github.com/nnsi/hono-practice --skill review-cycle
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: review-cycle
Source: https://github.com/nnsi/hono-practice/tree/main/.claude/skills/review-cycle
Command: npx skills add https://github.com/nnsi/hono-practice --skill review-cycle

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill orchestrates a dual-review cycle by running a sub-agent (Explore-type) and Codex in parallel, collecting their feedback, and iterating fixes until both return LGTM.

Core Features & Use Cases

  • Parallel dual-review: run architecture/design-focused feedback and implementation feasibility feedback concurrently.
  • Feedback integration & iterative fixes: merge inputs from both reviewers and apply changes, then re-run reviews.
  • Phase-based workflow: Phase 1 validate targets, Phase 2 execute parallel reviews, Phase 3 compile fixes, Phase 4 re-review until acceptance.
  • Robust guardrails: verify target directories exist and prevent over-flow of edits by enforcing exclusive-file grouping during parallel fixes.
  • Use cases: codebase plan validation for large repos, architecture validation during refactors, and improving PR quality with multi-perspective reviews.

Quick Start

Initiate Phase 1 to identify review targets, then run Phase 2 in parallel for Reviewer A and Reviewer B, and loop back through Phase 3 until LGTM from both.

Frequently Asked Questions about review-cycle

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

FAQPage Schema
How do I automate parallel code reviews for architecture and implementation feasibility?▼

To automate parallel code reviews, you can orchestrate concurrent feedback from multiple perspectives—such as architecture validation and implementation feasibility—iterating fixes until both reviewers return LGTM.

What is the best way to run dual code reviews until both reviewers approve?▼

Running dual code reviews requires a four-phase workflow: validate targets, execute parallel reviews, compile and apply integrated fixes, and re-review iteratively until both reviewers grant LGTM.

Can I use parallel review workflows for large codebase refactoring validation?▼

Yes, parallel review workflows support large codebase plan validation and architecture validation during refactors by concurrently evaluating design and implementation feasibility feedback.

How do code review orchestration tools prevent edit conflicts during parallel fixes?▼

Code review orchestration prevents edit conflicts during parallel fixes by enforcing exclusive-file grouping, ensuring that concurrent automated edits do not over-flow or modify the same files.

Do I need specific directory structures to run automated target discovery for code reviews?▼

Automated target discovery requires valid target directories; the workflow includes robust guardrails that verify target directories exist before initiating parallel reviewer orchestration and feedback integration.

Why use iterative re-review cycles instead of a single code review pass?▼

Iterative re-review cycles merge feedback from multiple reviewers, apply changes, and re-run evaluations to ensure both architecture and implementation perspectives achieve LGTM, improving overall PR quality.

Related Skills

Explore More Software Engineering Skills→
Nickalus12Nickalus12
2

perf-check

Identify performance hotspots and regressions in Loom-style codebases.

Community
Intermediate
thiagoooo34252thiagoooo34252
1

judgment-day

Launch parallel adversarial code reviews with two independent sub-agents.

Community
Advanced
vamseeachantavamseeachanta
2

Codex Review Workflow Skill

Automates AI-assisted code reviews for Claude Code and Google Gemini commits via Codex CLI.

Community
Advanced
ahiendzahiendz
3

lint-and-validate

Lint and statically analyze Node.js/TypeScript and Python codebases.

Community
Intermediate