ce:review

Analyze git diffs and generate tiered code review reports.

Updated Nov 11, 2023
One-click install
npx skills add https://github.com/michaelvolz/redmuffin.Blazor.StaticWeb --skill ce-review-michaelvolz
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ce:review
Source: https://github.com/michaelvolz/redmuffin.Blazor.StaticWeb/tree/main/.opencode/skills/ce/ce-review
Command: npx skills add https://github.com/michaelvolz/redmuffin.Blazor.StaticWeb --skill ce-review-michaelvolz

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Orchestrates a structured, multi-persona code review workflow so that every diff receives correctness, testing, maintainability, and standards scrutiny before any PR is opened or merged.

Core Features & Use Cases

  • Tiered persona coverage: Spawns always-on reviewers plus CE-specific agents and optional conditionals (security, performance, stack-specific reviewers, migration detectors) based on the diff, with protected artifacts kept untouched.
  • Mode-aware execution: Parses tokens like mode:autofix, mode:report-only, mode:headless, base:<ref>, and plan:<path> to enforce the appropriate interaction, autofix scope, artifactory rules, and headless constraints.
  • Scope, intent, and plan discovery: Resolves diff scopes via base detection scripts, summarizes intent, tracks plan requirements, and documents run artifacts, todos, schema drift checks, learnings, and deployment notes.

Quick Start

Invoke ce:review with the desired mode flags before opening a pull request.

Frequently Asked Questions about ce:review

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

FAQPage Schema
How do I run an automated code review on a git diff before opening a pull request?

Automated code review on git diffs is executed by analyzing changes to deliver a structured tiered report. It supports interactive, autofix, report-only, and headless modes to enforce appropriate interaction and constraints before opening a PR.

Can I automatically fix code issues found during a review without manual intervention?

Autofixing code issues is supported via the mode:autofix token, which enforces the appropriate autofix scope and artifactory rules. This allows the review process to generate fixes automatically while keeping protected artifacts untouched.

How does multi-persona code review work for checking maintainability and correctness?

Multi-persona code review works by spawning always-on reviewers alongside conditional agents for security, performance, and stack-specific checks based on the diff. It merges these persona findings to provide comprehensive correctness, testing, and maintainability scrutiny.

Does headless mode support running code reviews on GitHub branches and pull requests?

Headless mode supports running code reviews on GitHub branch or PR targets by applying headless constraints. It parses mode:headless tokens alongside base:<ref> and plan:<path> to enforce automation rules and safe execution without interactive input.

What is the best way to track plan requirements and schema drift during a code review?

Tracking plan requirements and schema drift during a code review is handled by resolving diff scopes, summarizing intent, and documenting run artifacts. It applies plan verification to track requirements, todos, learnings, and deployment notes.

When should I use report-only mode instead of interactive code review?

Report-only mode should be used when you need to generate a structured tiered code review report without applying fixes or interacting. It parses the mode:report-only token to enforce non-destructive analysis and artifact guidance for safe automation.