review

Review pre-merge code changes and emit structured findings.

1|Updated Apr 18, 2026
One-click install
npx skills add https://github.com/mopeyjellyfish/flywheel --skill review-mopeyjellyfish
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review
Source: https://github.com/mopeyjellyfish/flywheel/tree/main/skills/review
Command: npx skills add https://github.com/mopeyjellyfish/flywheel --skill review-mopeyjellyfish

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Code review is often slow, ad-hoc, and inconsistent. This Skill provides a structured, risk-first workflow to surface bugs, regressions, missing tests, and readiness gaps before merging.

Core Features & Use Cases

  • Structured reviewer orchestration using a central registry and optional stack packs to select targeted reviewers based on diff evidence.
  • Mode-driven interaction: interactive, autofix, report-only, and headless for automation and CI workflows.
  • Clear handoff and artifacts: outputs compact findings and optional full artifact payloads, run artifacts, and project standards traces.
  • Reusable, generic workflow that can be attached to any repository to speed up future PR reviews.

Quick Start

Run a Flywheel pre-merge code review on the current branch to surface structured findings.

Frequently Asked Questions about review

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

FAQPage Schema
How do I automate pre-merge code reviews to catch bugs and missing tests?

Automated pre-merge code review surfaces bugs, regressions, and missing tests by analyzing diff evidence. You can run it in headless or report-only mode within CI workflows to evaluate pull requests and branches before merging.

What is the best way to structure code reviews for consistency across multiple pull requests?

Structured code review uses a central registry and optional stack packs to deterministically select reviewers based on diff evidence. This enforces consistent, risk-first evaluation across all PRs and standalone branches.

Can I run code review workflows in CI pipelines without manual interaction?

Yes, code review workflows support headless and report-only modes designed for CI automation. These modes evaluate branches and emit structured compact findings and artifacts for downstream consumption without manual interaction.

How does diff evidence drive reviewer selection in a code review process?

Diff evidence drives reviewer selection by matching code changes against a central registry and stack packs. This deterministic loading process targets specific reviewers based on the actual files and technologies modified in a pull request.

Does code review automation work on standalone branches outside of pull requests?

Yes, code review automation applies to both pull requests and standalone branches. It evaluates the diff evidence to surface readiness gaps and emit structured findings regardless of the branch context.

What formats do structured code review findings output for downstream consumption?

Structured code review findings output compact reports and optional full artifact payloads. These include run artifacts and project standards traces designed for downstream consumption in CI workflows and automation systems.