reflection-loop

Automate self-critique and revision cycles to verify AI outputs.

226|55|Updated Mar 19, 2026
One-click install
npx skills add https://github.com/Miosa-osa/canopy --skill reflection-loop
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: reflection-loop
Source: https://github.com/Miosa-osa/canopy/tree/main/library/skills/ai-patterns/reflection
Command: npx skills add https://github.com/Miosa-osa/canopy --skill reflection-loop

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Enables AI systems to autonomously detect and correct mistakes by applying a structured critique loop, improving accuracy and reliability.

Core Features & Use Cases

  • Self-critique workflow: generates an initial response, critiques it, and applies targeted fixes.
  • Quality assurance: reduces errors in complex tasks, coding, or data reasoning.
  • Use Case: when a model produces uncertain results, run the reflection loop to identify and fix issues before final delivery.

Quick Start

Initiate the reflection loop on a task to automatically critique, revise, and verify the final answer.

Frequently Asked Questions about reflection-loop

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

FAQPage Schema
How do I make AI self-critique and fix its own output errors?

You can automate AI self-critique by running a reflection loop, which generates an initial response, identifies errors through structured critique, applies targeted fixes, and verifies the final deliverable against requirements.

What is an automated reflection loop for quality assurance?

An automated reflection loop is a self-correcting critique cycle for quality assurance that autonomously detects mistakes, proposes targeted revisions, and verifies correctness to improve AI output reliability before final delivery.

When should I use a self-critique workflow for AI-generated code?

Use a self-critique workflow for AI-generated code when handling complex programming tasks or data reasoning where errors and edge cases are likely, ensuring the model critiques and verifies uncertain results before final delivery.

How do I identify and resolve edge cases in AI programming outputs?

To identify and resolve edge cases in AI programming outputs, apply a structured critique loop that evaluates the initial response, detects issues, proposes targeted fixes, and verifies the final code meets all requirements.

Can I run a reflection loop on decision-support tasks with uncertain results?

Yes, you can run a reflection loop on decision-support tasks with uncertain results to automatically critique initial reasoning, identify logical issues, apply targeted fixes, and verify the final outcome meets requirements.

What are the limitations of using an automated self-critique cycle?

An automated self-critique cycle requires structured critique, issue identification, proposed fixes, and verification steps, meaning it demands sufficient processing depth to effectively review and revise complex task deliverables.