aifollowup

Generate three follow-up questions clarifying design, implementation, and bottleneck risks from unstaged changes and commits.

20|1|Updated Mar 20, 2026
One-click install
npx skills add https://github.com/radareorg/radare2-skills --skill aifollowup
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: aifollowup
Source: https://github.com/radareorg/radare2-skills/tree/main/dev/aifollowup
Command: npx skills add https://github.com/radareorg/radare2-skills --skill aifollowup

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Planning tasks and reviewing code often misses important issues that only become clear by extending the reasoning from the current unstaged changes and recent commits.

Core Features & Use Cases

  • Contextual follow-up questioning: Produces targeted questions that help surface design tradeoffs, implementation risks, bottlenecks, and overlooked constraints.
  • Workflow for ongoing work: Focuses on non-master branch commits, unstaged changes, and the todo roadmap so follow-ups align with what’s actually happening next.
  • Use Case: After making several related refactors, use this skill to quickly identify where the architecture, performance, or correctness assumptions may be breaking down as you move forward.

Quick Start

Ask your agent to run aifollowup to extend reasoning from your current unstaged changes and commits, then return three FOLLOW-UP QUESTIONS.

Frequently Asked Questions about aifollowup

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

FAQPage Schema
How do I generate follow-up questions for code review and task planning?

Generate follow-up questions by cross-considering unstaged changes, current non-master commits, and the todo roadmap to produce exactly three concise questions oriented around design issues, implementation problems, and bottleneck investigation.

What are follow-up questions in software design used for?

Follow-up questions in software design are used to surface design tradeoffs, implementation risks, and overlooked constraints that only become clear by extending reasoning from current unstaged changes and recent commits during active development work.

How do I identify implementation risks and performance bottlenecks in ongoing work?

Identify implementation risks and performance bottlenecks by analyzing non-master branch commits and unstaged changes against the todo roadmap, producing targeted questions that reveal where architecture, performance, or correctness assumptions may break down.

Can I use commit analysis to check if my refactoring breaks architecture assumptions?

Yes, you can use commit analysis to check if your refactoring breaks architecture assumptions by extending reasoning from your recent commits and unstaged changes to quickly identify where correctness assumptions are failing as you move forward.

Does follow-up question generation work with unstaged changes and non-master branch commits?

Yes, follow-up question generation works specifically with unstaged changes and non-master branch commits, cross-considering these elements alongside the todo roadmap so that the resulting questions align with what is actually happening next in your workflow.

What is the best way to surface design tradeoffs during active development?

The best way to surface design tradeoffs during active development is to generate exactly three concise follow-up questions oriented around design issues, implementation problems, and bottleneck investigation by cross-considering your current commits and todo roadmap.