triple-review

Coordinate multi-agent adversarial planning, smoke testing, and gated code reviews for refactoring workflows.

8|2|Updated Feb 8, 2026
One-click install
npx skills add https://github.com/psaab/xpf --skill triple-review-psaab
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: triple-review
Source: https://github.com/psaab/xpf/tree/main/.claude/skills/triple-review
Command: npx skills add https://github.com/psaab/xpf --skill triple-review-psaab

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill prevents architectural drift and performance regressions in high-performance networking code by enforcing a mandatory, multi-agent adversarial review process before any code is merged.

Core Features & Use Cases

  • Adversarial Planning: Coordinates parallel plan reviews by Codex and Gemini to catch architectural flaws before implementation begins.
  • Automated Smoke Testing: Executes a comprehensive matrix of v4/v6, push/reverse, and CoS-enabled/disabled tests to ensure performance stability.
  • Use Case: When refactoring the userspace dataplane, use this Skill to ensure that every change is planned, reviewed by three independent AI agents, and verified against line-rate performance requirements.

Quick Start

Invoke the triple-review skill with the issue number and a short scope description to initialize the refactor worktree and begin the planning phase.

Frequently Asked Questions about triple-review

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

FAQPage Schema
How do I prevent architectural drift when refactoring high-performance networking code?

Prevent architectural drift during networking code refactoring by enforcing a multi-agent adversarial review process before any code is merged. This coordinates parallel plan reviews by independent AI agents, ensuring architectural consistency and preventing performance regressions.

What is multi-agent adversarial planning for code refactoring?

Multi-agent adversarial planning for code refactoring is a process where independent AI agents, such as Codex and Gemini, conduct parallel plan reviews. This technique catches architectural flaws and enforces consensus-based merge gates before implementation begins.

How do I automate smoke testing for v4/v6 and push/reverse networking performance?

Automate smoke testing for v4/v6 and push/reverse networking performance by executing a comprehensive test matrix. This includes CoS-enabled and disabled configurations to verify line-rate performance stability and prevent regressions.

Does multi-agent code review work for userspace dataplane refactoring?

Multi-agent code review works for userspace dataplane refactoring by verifying every change against line-rate performance requirements. It ensures modifications are planned, reviewed by three independent AI agents, and gated for architectural consistency.

How do I enforce bisectable commit history during complex systems engineering tasks?

Enforce bisectable commit history during complex systems engineering tasks by applying gated code reviews and rigorous multi-agent adversarial planning. This process ensures multi-stack performance verification and consensus-based merges for architectural consistency.