AI Avatar Critique

Ruthlessly critique an AI video avatar before it ships, with a verdict, scored dimensions and a timestamped fix list. Use when the user says 'critique this avatar', 'review my avatar video', 'avatar QC', or asks why an avatar looks fake.

39 stars 5 forksAI, LLMs & Agents Plugin marketplace

Install

Claude Code plugin
/plugin marketplace add lionelsimai/claude-skills-collection

The repo is a Claude Code plugin marketplace. Run /plugin afterwards to pick the plugin that includes this skill.

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add lionelsimai/claude-skills-collection --skill ai-avatar-critique
Manual install
git clone --depth 1 https://github.com/lionelsimai/claude-skills-collection.git /tmp/claude-skills-collection
mkdir -p ~/.claude/skills
cp -r /tmp/claude-skills-collection/skills/ai-avatar-critique ~/.claude/skills/ai-avatar-critique

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# AI Avatar Critique — world-class self-critique for AI video avatars

## Purpose

- Judge an AI avatar video the way a hostile expert, a scrolling viewer and the person whose face it is would judge it, all in one pass.

- Never let a video ship on technical checks alone. A clean ffprobe, the right duration and a loudness pass say nothing about whether the avatar looks alive, sounds like the person or holds attention.

- Produce one verdict, a scored critique, a timestamped fix list ranked by severity, and a re-score after fixes. No vague praise and no unranked notes.

## When to use it

- Use it on every avatar video before delivery or posting, whether it came from HeyGen, Sogni, HyperFrames, Synthesia, D-ID, Hedra, a talking-head pipeline or a real-person edit with an AI voice.

- Use it on a competitor or reference avatar when the user asks what makes it work or fail.

- Use it again after every fix round. A critique without a re-score is unfinished work.

## Verdict first, always

- Lead with one verdict in the first line. SHIP means it can go out as is. FIX means named fixes would make it shippable. KILL means do not post it and do not patch it, rebuild it.

- Follow the verdict with the total score out of 100, the weakest dimension, and the single worst moment with its timestamp.

- Give depth after the verdict, not before it. the user wants the call first, then the evidence.

## Hard fails — any one of these is an automatic KILL
Read the full skill on GitHub

Details

Skill name
ai-avatar-critique
Path
skills/ai-avatar-critique/SKILL.md
License
MIT
Repo updated
8 November 2026

FAQ

What does the AI Avatar Critique skill do?

Ruthlessly critique an AI video avatar before it ships, with a verdict, scored dimensions and a timestamped fix list. Use when the user says 'critique this avatar', 'review my avatar video', 'avatar QC', or asks why an avatar looks fake.

How do I install the AI Avatar Critique skill?

Claude Code plugin: /plugin marketplace add lionelsimai/claude-skills-collection. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is AI Avatar Critique popular?

It lives in lionelsimai/claude-skills-collection, which has 39 GitHub stars and 5 forks.

Related skills

1031 Exchange Guide

39

lionelsimai/claude-skills-collection

Guide 1031 exchange processes and requirements. TRIGGERS - Use when user needs help with 1031-exchange-guide related tasks.

Productivity & Notes

30 60 90 Plan

39

lionelsimai/claude-skills-collection

Create 30-60-90 day plans for new hires. TRIGGERS - Use when user needs help with 30-60-90-plan related tasks.

Productivity & Notes

360 Feedback

39

lionelsimai/claude-skills-collection

Design 360-degree feedback programs with surveys. TRIGGERS - Use when user needs help with 360-feedback related tasks.

Productivity & Notes

Alterlab Nmc Digital Ethics

21

AlterLab-IEU/AlterLab-FC-Skills

This skill should be used when the user asks about "digital ethics", "media ethics", "AI ethics", "platform governance", "misinformation", "disinformation analysis", "act as a digital ethics advisor", "digital ethics mode", "algorithmic bias", "data privacy", "content moderation", "ethical framework", "deepfake", "surveillance", "fact-checking tools", "verification workflow", "platform audit", "ethics stress test", "algorithmic accountability", "informed consent", "digital rights", "tech regulation", or needs expertise in analyzing ethical dilemmas in digital media, AI, and platform ecosystems

AI, LLMs & AgentsIn 1 list

Agent Setup Service

39

lionelsimai/claude-skills-collection

Run an always-on AI agent setup service for owner-led businesses, with rulebooks and a client setup pack. Use when the user asks to set up AI agents for a business, chase invoices with AI, or auto-reply to leads.

AI, LLMs & Agents

Zendesk

8

andmarios/zendesk-skill

Interact with Zendesk Support via CLI - search tickets, view details, analyze metrics, manage users/organizations, and update tickets. All responses are saved locally for efficient jq querying. All Zendesk content is screened for prompt injection (regex, semantic, and LLM-based) and wrapped with security markers before reaching the LLM.

AI, LLMs & Agents

Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.