review

Coordinate parallel specialist code reviews and aggregate findings into unified reports.

15|1|Updated Mar 29, 2026
One-click install
npx skills add https://github.com/duronext/jig --skill review-duronext
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review
Source: https://github.com/duronext/jig/tree/main/core/skills/review
Command: npx skills add https://github.com/duronext/jig --skill review-duronext

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Coordinating code reviews across multiple specialists can be slow and brittle; this Skill streamlines pre-PR and quality-gate reviews by dispatching a swarm of focused reviewers and aggregating their findings into a single report.

Core Features & Use Cases

  • Dispatches parallel specialist reviewers (security, correctness, performance, and more) to cover multiple concerns concurrently.
  • Integrates with code-review, team-dev, and /review invocations to enforce thorough, consistent reviews during PRs and gate checks.
  • Produces a unified swarm report with actionable findings, enabling faster decisions and clearer remediation guidance.

Quick Start

Invoke /review during a code review to start the swarm-driven evaluation.

Frequently Asked Questions about review

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

FAQPage Schema
How do I run parallel code reviews with a swarm of specialists?

Parallel code reviews with a swarm of specialists are dispatched by invoking /review during your pre-PR verification. The swarm concurrently evaluates security, correctness, and performance, then aggregates findings into a single unified report to guide remediation.

What is configuration-driven discovery for swarm specialist reviewers?

Configuration-driven discovery is the mechanism that automatically identifies and selects tier-based specialist reviewers for the code-review swarm. It enforces consistent participation across focused domains and outputs a unified set of actionable findings.

Can I integrate automated code review into team-dev quality gates?

Yes, automated code review integrates directly into team-dev quality gates and pre-PR checks. Invoking /review dispatches the specialist swarm to enforce thorough, consistent verification and produces a unified report for faster decisions.

Does parallel review work without external dependencies or components?

Yes, parallel review operates with zero external dependencies or components. It internally manages the configuration-driven discovery, tier-based participation, and automated aggregation required to dispatch the specialist swarm and report findings.

What is the best way to aggregate code review findings from multiple specialists?

The best way to aggregate code review findings is dispatching a parallel specialist swarm via /review. It automates the aggregation of security, correctness, and performance findings into a single unified report, accelerating pre-PR remediation decisions.

When should I not use a parallel swarm for pre-PR code review?

You should avoid a parallel swarm for pre-PR code review if your changes are trivial and do not warrant multi-specialist verification across security, correctness, and performance, as dispatching tier-based reviewers adds overhead for minimal remediation value.