Screenshot Critic — Data & AI agent for Claude Code
Independent design critic that judges a UI from screenshots only, never code, history, or rationale.
How to install Screenshot Critic
Installs to ~/.claude/agents/cerisecloud-design-studio-skills-screenshot-critic.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/cerisecloud/design-studio-skills/HEAD/agents/screenshot-critic.md -o ~/.claude/agents/cerisecloud-design-studio-skills-screenshot-critic.md Restart Claude Code, or start a new session, for it to be picked up.
What Screenshot Critic does
name: screenshot-critic description: Independent design critic that judges a UI from screenshots only, never code, history, or rationale. Returns the aesthetic it reads, the biggest gaps against a top-studio execution, AI tells, what to remove, and a score out of 10. Used by the design-critic and design-studio skills; can also be invoked directly to critique any screenshots. tools: Read model: inherit
You are the design director at a top studio, reviewing work before it goes to the clie
Alternatives in Data & AI
- LLM AI Hunter — LLM and Agentic AI vulnerability specialist 812 ★
- AI Maestro: A Development History — From "AI Agents Web Dashboard" to multi-host agent orchestration in 110 days 551 ★
- Patina Naturalness Reviewer — Triggers to re-scan a patina rewrite for residual AI tells and over-editing risk 352 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Noemeo UX Designer
Independent UI/UX review of a Noemeo interface -- page, component, flow, plugin dialog, screenshot or mock --
Logo Critic
Blind, independent scorer for logo candidates, especially AI-generated ones. Use after any generation round or
Payments Reviewer
Reviews Rails code that moves money for idempotency, database-level uniqueness, money types, append-only histo
AI Image Director
Desk Studio's AI Image Director. Directs generated imagery like a photographer: whether it should exist, the s
Studio Director
Phase-aware Forge AI Studio orchestrator. Splits one approved phase goal into independent workstreams, delegat
Bm Framework Selector
Presents an interactive decision matrix to surface the right AI/LLM framework for the user's specific use case
Related Skills
Orchestrate Consult
Run an independent external audit of a completed ZOdyssey run. Hands the plan + full git diff to the external
Leash
Jev-powered guardrail that keeps a coding agent's output quality on a short leash: judges every turn against y
Aesthetic Blanks
Isolated, independent aesthetic website templates (parts-clones). Each site is a self-contained Next.js wirefr