gsd-review

Review phase plans by invoking external AI CLIs and aggregating critiques into REVIEWS.md.

Updated May 21, 2026

Search Tags:#code review #documentation #phase planning #prompting #ai cli #multi-model feedback

Ready to run

npx skills add https://github.com/jedmamosto/m-and-ms --skill gsd-review-jedmamosto

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: gsd-review
Source: https://github.com/jedmamosto/m-and-ms/tree/main/.agents/skills/gsd-review
Command: npx skills add https://github.com/jedmamosto/m-and-ms --skill gsd-review-jedmamosto

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you validate and improve phase plans by getting independent feedback from multiple external AI coding tools before you commit to execution.

Core Features & Use Cases

Cross-AI peer review orchestration: Runs reviewer prompts across selectable external AI CLIs (Gemini, Claude, Codex, OpenCode, Qwen Code, Cursor).
Phase-scoped review: Targets a specific phase number extracted from the required argument so feedback stays actionable and bounded.
Structured output for iteration: Collects all reviewer responses into a single REVIEWS.md file suitable for feeding back into planning workflows.

Quick Start

Ask the system to review phase 3 by running gsd-review with the required argument for the phase number.

Frequently Asked Questions about gsd-review

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

FAQPage Schema

How do I get multi-model feedback on a phase plan before execution?▼

To get multi-model feedback on a phase plan, you can run cross-AI peer review orchestration that invokes multiple external AI CLIs and aggregates their critiques into a consolidated REVIEWS.md file.

What AI CLIs can I use for cross-AI code review and phase planning validation?▼

You can use selectable external AI CLIs including Gemini, Claude, Codex, OpenCode, Qwen Code, and Cursor to run reviewer prompts and collect independent validation for phase planning.

How do I target a specific phase number when reviewing an implementation plan?▼

You target a specific phase by providing the phase number as a required argument, which scopes the review so feedback stays actionable and bounded to that specific phase.

Does gsd-review require any specific dependencies to aggregate AI critiques?▼

gsd-review requires no installed dependencies, but it needs CLI detection for the selected external AI coding tools to execute per-reviewer prompting and collect responses successfully.

What is the best way to consolidate AI critiques for downstream planning workflows?▼

The best way to consolidate AI critiques is by collecting all reviewer responses into a single structured REVIEWS.md artifact suitable for feeding back into planning and delivery workflows.

Can I use multiple AI coding tools to validate a single phase plan independently?▼

Yes, you can validate a single phase plan independently by running reviewer prompts across different model providers and toolchains, aggregating their distinct critiques into one file.