skills.rest
calcosmic
calcosmicCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

cross-model-plan-review

Dispatch colony plans to external AI CLIs for independent review and convergence reporting.

9|1|Updated Feb 1, 2026
Search Tags:#convergence#cross-model#structured-feedback#plan-review#external-review#independent-review#ai-clis
One-click install
npx skills add https://github.com/calcosmic/Aether --skill cross-model-plan-review
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: cross-model-plan-review
Source: https://github.com/calcosmic/Aether/tree/main/.aether/skills-codex/colony/cross-model-plan-review
Command: npx skills add https://github.com/calcosmic/Aether --skill cross-model-plan-review

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

No single AI sees every blind spot. This skill dispatches colony plans to external AI CLIs (Claude, GPT, Gemini, etc.) for independent review, then converges their feedback into a unified assessment.

Core Features & Use Cases

  • Independent reviews from multiple AI CLIs surface diverse perspectives.
  • Convergence analysis merges feedback into a cohesive verdict with consensus and unique concerns.
  • Structured feedback output includes reviewer concerns, severity, and an overall recommendation.
  • Re-planning loop allows up to three cycles when high-severity issues arise.
  • Ideal for validating high-risk plans before milestones and for architectural quality gates.

Quick Start

Submit the plan to available external AI CLIs for independent review and generate a convergence report.

Frequently Asked Questions about cross-model-plan-review

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

FAQPage Schema
How does cross-model plan review work for catching blind spots?▼

Yes, you can validate high-risk plans through independent cross-model reviews before milestones. The workflow dispatches plans to multiple external AI CLIs to gather diverse assessments, ensuring safer decisions and architectural quality gates.

What is convergence reporting in AI plan evaluation?▼

Convergence reporting merges independent feedback from multiple AI CLIs into a cohesive verdict. It identifies consensus concerns, unique issues, and severity levels, providing an overall recommendation for safer plan validation.

How do I handle high-severity issues found during external AI plan reviews?▼

You handle high-severity issues using the re-planning loop, which allows up to three cycles. This iterative process dispatches revised plans back to external AI CLIs until severe blind spots are resolved and convergence is achieved.

Does cross-model review require specific AI CLIs to generate structured feedback?▼

No specific CLI is mandated, but it dispatches plans to available external AI CLIs like Claude, GPT, and Gemini. The process generates structured feedback outputs containing reviewer concerns, severity, and overall recommendations.

What are the limitations of using independent reviews for architectural quality gates?▼

A limitation of independent reviews is the reliance on available external AI CLIs to surface blind spots. The re-planning loop is capped at three cycles to resolve high-severity issues, which may not exhaust all unique concerns.

Related Skills

Explore More Product & Management Skills→
TruCol270TruCol270
1

gsd-review

Coordinate cross-AI peer reviews of phase plans from external AI CLIs.

Community
Advanced
deerawandeerawan
1

gsd-plan-review-convergence

Automate cross-AI plan review loops until HIGH-severity concerns are resolved.

Community
Advanced
masharrattmasharratt
14

cfn-loop-output-processing

Consolidate CFN Loop agent outputs into structured Loop3Result and Loop2Result JSON.

Community
Advanced
salverius-techsalverius-tech

structured-analysis

Apply structured analytical frameworks to evaluate prompts, systems, documents, and code.

Community
Advanced