Self-Review

Run a 5-point checklist to verify AI responses before sending.

Updated Mar 11, 2026
One-click install
npx skills add https://github.com/aggelosChatziioannou/kimi_claw_skills --skill self-review-aggeloschatziioannou
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Self-Review
Source: https://github.com/aggelosChatziioannou/kimi_claw_skills/tree/main/self-review
Command: npx skills add https://github.com/aggelosChatziioannou/kimi_claw_skills --skill self-review-aggeloschatziioannou

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires self-review.sh, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill acts as a crucial quality control layer, ensuring that all AI-generated responses are accurate, complete, appropriate, and safe before they are sent to the user.

Core Features & Use Cases

  • 5-Point Checklist: Verifies correctness, completeness, tone, fact verification, and safety.
  • Contextual Application: Applies different review strictness based on the conversation's context (e.g., technical instructions vs. casual chat).
  • Use Case: Before sending a code snippet or a critical system status update, this Skill runs a rapid check to catch potential errors or misrepresentations, preventing user confusion or system issues.

Quick Start

Run the self-review script on the proposed response text.

Frequently Asked Questions about Self-Review

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

FAQPage Schema
How do I ensure quality control and accuracy for AI-generated responses before sending?

You ensure quality control for AI-generated responses by applying a systematic 5-point checklist that verifies correctness, completeness, tone, fact verification, and safety prior to output.

What is response validation and how does a pre-send safety check work?

Response validation is a pre-send safety check mechanism that reviews generated text against defined criteria to prevent user confusion or system issues by catching potential errors or misrepresentations.

How do I run a self-review on a proposed AI response or code snippet?

You run a self-review by executing the self-review script directly on the proposed response text to rapidly catch potential errors before the message is delivered to the user.

Can I adjust the response validation strictness for casual chat versus technical instructions?

Yes, you can adjust the review strictness because the skill applies contextual application, meaning it scales its verification rigor based on whether the context involves technical instructions or casual chat.

What is the best way to prevent errors in critical system status updates generated by AI?

The best way to prevent errors in critical updates is to implement a pre-send verification process that performs a rapid fact verification and safety check on the proposed text.

What are the limitations of using a checklist for response validation in software engineering?

The limitation of using a checklist for response validation is that its effectiveness depends on contextual application, meaning a basic checklist might not catch deep logic errors without adjustable rigor for complex technical instructions.