skills.rest
Zeptiny
ZeptinyCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

ce-code-review

Automate GitHub code reviews with persona agents and CI/CD integration.

Updated Apr 28, 2026
Search Tags:#quality assurance#code review#automated review#issue detection#agent-based
One-click install
npx skills add https://github.com/Zeptiny/opencode-setup --skill ce-code-review-zeptiny
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ce-code-review
Source: https://github.com/Zeptiny/opencode-setup/tree/main/skills/ce-code-review
Command: npx skills add https://github.com/Zeptiny/opencode-setup --skill ce-code-review-zeptiny

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

The ce-code-review skill provides a comprehensive, automated code review process, identifying and addressing potential issues early in the development cycle, ensuring code quality and maintainability.

Core Features & Use Cases

  • Structured Code Review: Utilizes tiered persona agents to deliver detailed, structured feedback on code changes.
  • Multi-Agent Collaboration: Collaborative requirements exploration and dialogue to ensure thorough code review.
  • Confidence-Gated Findings: Ensures that only high-confidence findings are presented for review.
  • Merge/Dedup Pipeline: Efficiently merges and deduplicates findings to streamline the review process.
  • Autofix Mode: Allows for the automatic application of safe fixes without user intervention.
  • Report-Only Mode: Generates a detailed report without making any changes to the codebase.
  • Headless Mode: Suitable for programmatic invocation, providing structured findings and run artifacts.

Quick Start

To initiate a code review, use the ce-code-review skill with the mode:autofix flag on the PR link: ce-code-review mode:autofix [PR link].

Frequently Asked Questions about ce-code-review

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

FAQPage Schema
How do I automate code review for pull requests to catch logic errors and security vulnerabilities?▼

Automated code review uses tiered persona agents to analyze pull requests, focusing on logic errors, security vulnerabilities, and performance bottlenecks. It requires access to the GitHub repository and PR metadata to provide structured feedback.

Can I run automated code review in a CI/CD pipeline without user intervention?▼

Yes, automated code review supports a headless mode for programmatic invocation in CI/CD pipelines and an autofix mode to apply safe fixes without user intervention. It also offers a report-only mode to generate findings without modifying the codebase.

What is the best way to ensure code review findings are accurate and not duplicated?▼

Automated code review uses confidence-gated findings to ensure only high-confidence issues are presented. A merge and dedup pipeline efficiently merges and deduplicates these findings to streamline the review process and reduce noise.

Does this automated code review tool work with Python-based tooling and GitHub repositories?▼

Yes, the automated code review process is compatible with Python-based tooling and integrates with GitHub. It requires access to the GitHub repository, PR metadata, and the codebase to identify issues and provide feedback.

How does multi-agent collaboration work for code quality assurance?▼

Multi-agent collaboration for code quality assurance uses a suite of specialized reviewer personas to explore requirements and dialogue. This structured approach ensures thorough code review across maintainability, performance, and project standards.

When should I use report-only mode instead of autofix mode for code review?▼

Use report-only mode when you need a detailed review report without modifying the codebase, ensuring safe observation. Use autofix mode when you want the automated review to directly apply safe fixes without user intervention.

Related Skills

Explore More Software Engineering Skills→
SpillwaveSolutionsSpillwaveSolutions
46

pr-reviewer

Collect GitHub PR data and generate structured review files with two-stage approval.

Official
Advanced
DieStokDieStok

ce-code-review

Orchestrate multi-agent code review workflows and aggregate findings into merge-ready reports.

Community
Advanced
All-The-VibesAll-The-Vibes
50

ce-review

Automate structured code review with tiered personas and parallel sub-agents.

Official
Advanced
softwaresaltsoftwaresalt
1

review

Coordinate structured code review across multiple reviewer personas for git diffs.

Community
Advanced