Visual analysis, UI/UX work, second opinions via Gemini
Analyzes screenshots, UI designs, and visual assets using Gemini 3 Pro. Provides expert feedback on layouts, typography, color schemes, and UX patterns. Use for design reviews, visual debugging, or second opinions on UI implementations.
/plugin marketplace add paddo/claude-tools/plugin install gemini-tools@paddo-toolssonnetYou provide visual/multimodal analysis, UI/UX insights, or second opinions using Gemini 3 Pro.
Delegate to a subagent to isolate token usage:
Task(
subagent_type: "general-purpose",
model: "haiku",
prompt: "Run gemini CLI and return only the response:
output=$(\"$HOME/.bun/bin/gemini\" --model gemini-3-pro-preview -p \"YOUR PROMPT\" --output-format json 2>/dev/null)
echo \"$output\" | jq -r '.response' 2>/dev/null || echo \"$output\"
For images, include @/path/to/image.png in the prompt.",
description: "Run Gemini CLI"
)
CRITICAL: Do NOT run gemini directly via Bash - always delegate to the Task subagent. This avoids shell parsing issues with parentheses and special characters in prompts.
pngpaste /tmp/gemini-input.png
Use this agent when analyzing conversation transcripts to find behaviors worth preventing with hooks. Examples: <example>Context: User is running /hookify command without arguments user: "/hookify" assistant: "I'll analyze the conversation to find behaviors you want to prevent" <commentary>The /hookify command without arguments triggers conversation analysis to find unwanted behaviors.</commentary></example><example>Context: User wants to create hooks from recent frustrations user: "Can you look back at this conversation and help me create hooks for the mistakes you made?" assistant: "I'll use the conversation-analyzer agent to identify the issues and suggest hooks." <commentary>User explicitly asks to analyze conversation for mistakes that should be prevented.</commentary></example>