gsd-plan-review-convergence

Orchestrate iterative plan-review-replan loops with multiple AI CLI reviewers.

1|Updated Apr 4, 2026
One-click install
npx skills add https://github.com/nnexai/git-stacks --skill gsd-plan-review-convergence-nnexai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: gsd-plan-review-convergence
Source: https://github.com/nnexai/git-stacks/tree/main/.codex/skills/gsd-plan-review-convergence
Command: npx skills add https://github.com/nnexai/git-stacks --skill gsd-plan-review-convergence-nnexai

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This skill solves the problem of fragmented and unverified project planning by automating the iterative review of plans across multiple AI models until all high-priority concerns are resolved.

Core Features & Use Cases

  • Cross-AI Convergence: Orchestrates a feedback loop between planning agents and external AI reviewers (Codex, Gemini, Claude, etc.) to ensure plan robustness.
  • Stall Detection & Escalation: Monitors review cycles to prevent infinite loops and triggers escalation gates when plans fail to converge.
  • Use Case: Use this when you have a complex project plan and need to ensure it meets rigorous quality standards by having multiple AI reviewers critique it before execution begins.

Quick Start

Invoke the convergence workflow for phase one by typing $gsd-plan-review-convergence 1 followed by your preferred reviewer flag.

Frequently Asked Questions about gsd-plan-review-convergence

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

FAQPage Schema
What is automated multi-agent plan review and convergence for software engineering?

Automated plan review convergence orchestrates an iterative feedback loop between planning agents and multiple AI CLI reviewers to ensure software project plans meet rigorous quality standards before execution begins.

How do I run an automated cross-AI plan review cycle for my project plan?

Invoke the convergence workflow by typing the command followed by your phase number and preferred reviewer flag, which triggers the automated iterative plan-review-replan loop using your selected AI CLI reviewers.

Can I use specific AI CLI reviewers like Codex, Gemini, or Claude to validate my project plans?

Yes, the workflow supports configurable reviewer selection, allowing you to apply external AI reviewers such as Codex, Gemini, and Claude to critique and validate complex software engineering project plans.

How does stall detection work during iterative AI plan review cycles?

Stall detection monitors the review cycles to prevent infinite feedback loops, automatically triggering escalation gates when project plans fail to converge and high-priority concerns remain unresolved.

When do I need multi-agent validation and convergence for my software engineering workflows?

You need multi-agent validation when handling complex software engineering workflows that require automated quality gating, structured plan verification, and robust cross-AI consensus before execution proceeds.

What happens if my project plan fails to converge during the automated review process?

If a plan fails to converge, stall detection prevents infinite loops and triggers escalation gates, halting the iterative review process to prevent execution of an unverified or fragmented project plan.