Reflection Skill

Run an isolated sub-agent to critique and validate significant outputs.

9|2|Updated Dec 10, 2025
One-click install
npx skills add https://github.com/ilandahan/AID --skill reflection-skill
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Reflection Skill
Source: https://github.com/ilandahan/AID/tree/main/.claude/skills/reflection
Command: npx skills add https://github.com/ilandahan/AID --skill reflection-skill

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides automatic quality verification by spinning up an isolated sub-agent to critique significant outputs, ensuring transparency and higher trust.

Core Features & Use Cases

  • Isolated evaluation via sub-agent to avoid context leakage and bias.
  • Visible Quality Check display showing results to users.
  • Works across deliverables: code, architecture, PRDs, specs, tests, APIs, and schemas.
  • Use case example: Imagine reviewing a complex PRD; the reflection skill can validate alignment and completeness before sign-off.

Quick Start

To activate, send /reflect after generating a draft or specification. The skill will extract the necessary context, spawn the reflection-agent, parse the results, and present the Quality Check box for review.

Frequently Asked Questions about Reflection Skill

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

FAQPage Schema
How do I automate quality checks for generated code and specifications?

Automate quality checks by running an isolated self-critique sub-agent that evaluates significant outputs like code, specs, and schemas. This prevents context bias and presents a structured Quality Check display for transparent validation of your deliverables.

How does an isolated sub-agent prevent bias during AI self-critique?

An isolated sub-agent prevents bias by extracting context and evaluating outputs independently from the main generation process. This separation avoids context leakage, ensuring the self-critique provides a transparent and objective quality assessment.

What's the best way to validate a PRD for completeness before sign-off?

Validate a PRD for completeness by triggering an isolated reflection sub-agent to critique the draft. It evaluates alignment and structure, parsing results into a JSON format to display a final Quality Check box for review before sign-off.

Can I use automated self-critique to review architecture and API schemas?

Yes, you can use automated self-critique to review architecture and API schemas. The reflection mechanism applies across diverse deliverables including code, tests, plans, and APIs, providing structured evaluation for each specific output type.

When should I run an isolated quality check on my generated outputs?

Run an isolated quality check when generating significant outputs like code, architecture, PRDs, specs, tests, plans, APIs, or schemas. Triggering the reflection process after drafting these deliverables ensures structured evaluation and higher trust before finalizing.