skills.rest
sommio
sommioCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

ce:review

Orchestrate multi-agent code reviews and merge findings into a single report.

1|Updated Apr 3, 2026
Search Tags:#automation#agents#synthesis#diff#code-review#pull-request
One-click install
npx skills add https://github.com/sommio/RSSift --skill ce-review-sommio
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ce:review
Source: https://github.com/sommio/RSSift/tree/main/.agents/skills/ce-review
Command: npx skills add https://github.com/sommio/RSSift --skill ce-review-sommio

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This skill orchestrates automated, structured code reviews by spawning multiple persona reviewers to produce a merged, deduplicated findings report for code changes before PRs.

Core Features & Use Cases

  • Spawns parallel reviewer personas and merges findings into a single report
  • Supports interactive, autofix, report-only, and headless modes
  • Integrates with PR metadata and diff context to guide intent-aware review

Quick Start

Invoke ce:review on the current branch or a PR to start a collaborative, automated code review.

Frequently Asked Questions about ce:review

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

FAQPage Schema
How do I automate code reviews for pull requests?▼

Automated code review is achieved by spawning multiple persona reviewers that analyze PR metadata and diff context, merging their findings into a single, deduplicated report. You can invoke the review process on a current branch or PR to start.

What is a multi-agent code review workflow?▼

A multi-agent code review workflow orchestrates specialized reviewer personas to analyze code changes in parallel, routing actions and surfacing structured findings with confidence levels before a pull request is merged.

Can I run automated code reviews in a headless CI pipeline?▼

Yes, automated code reviews support headless, report-only, autofix, and interactive modes. Headless mode allows you to integrate the multi-agent review pipeline directly into CI for scalable, policy-compliant checks.

How do I deduplicate findings from multiple code review agents?▼

Findings from multiple review agents are deduplicated by merging the specialized persona results into a single structured report. This process uses diff context to ensure action routing and confidence levels remain accurate.

Does automated code review work with PR metadata and diff context?▼

Yes, the review process integrates directly with PR metadata and diff context to guide intent-aware analysis. This ensures the generated findings report accurately reflects the specific changes and intent of the pull request.

Related Skills

Explore More Software Engineering Skills→
edobryedobry
6

review-pr

Analyze pull requests and post structured GitHub PR reviews.

Community
Advanced
Dhruvipatel0514Dhruvipatel0514
1

code-auditor

Identify and fix OWASP security vulnerabilities in source code repositories.

Community
Intermediate
yxc023yxc023

ce:review

Coordinate multi-agent code review workflows and generate structured findings.

Community
Advanced
vitalliumvitallium
10

ce-code-review

Orchestrate multiple reviewer personas to surface structured findings from a diff.

Community
Advanced