Skill

gemini-prompting

Internal guidance for composing effective Gemini prompts for visual analysis, design review, and coding tasks

Install

Run in your terminal

npx claudepluginhub markbrutx/gemini-plugin-cc

Tool Access

This skill uses the workspace's default tool permissions.

Skill Content

Similar Skills

cache-components

Guides Next.js Cache Components and Partial Prerendering (PPR) with cacheComponents enabled. Implements 'use cache', cacheLife(), cacheTag(), revalidateTag(), static/dynamic optimization, and cache debugging.

cache-components

138.6k

claude-opus-4-5-migration

2 files

Migrates code, prompts, and API calls from Claude Sonnet 4.0/4.5 or Opus 4.1 to Opus 4.5, updating model strings on Anthropic, AWS, GCP, Azure platforms.

claude-opus-4-5-migration

83.2k

evaluation-methodology

1 file

Details PluginEval's skill quality evaluation: 3 layers (static, LLM judge), 10 dimensions, rubrics, formulas, anti-patterns, badges. Use to interpret scores, improve triggering, calibrate thresholds.

plugin-eval

32.9k

Stats

Stars1

Forks0

Last CommitApr 3, 2026

Actions

View Source View Plugin View on GitHub View README

<context> Brief project context and what the user is working on. </context> <task> Clear, specific instruction. One task per prompt. </task> <constraints> Output format, style guidelines, framework requirements. </constraints>

Anti-Patterns

Do not dump entire codebases into the prompt — Gemini works best with focused context

Do not ask Gemini to reason about complex logic — that's Claude's strength

Do not ask Gemini to write backend/infrastructure code — delegate only visual/design tasks

Do not include prompts longer than ~4000 words — keep it focused

Gemini Prompting Guide

Gemini excels at two things that Claude does not:

Visual analysis — screenshots, mockups, UI layouts, design comparison
Design/styling — color palettes, modern frontend aesthetics, creative UI

Prompt Structure

Good Gemini prompts follow this pattern:

<context>
Brief project context and what the user is working on.
</context>

<task>
Clear, specific instruction. One task per prompt.
</task>

<constraints>
Output format, style guidelines, framework requirements.
</constraints>

Visual Analysis Prompts

When the task involves visual input:

Describe what the screenshot/mockup shows
Be specific about what to look for (spacing, color, alignment, hierarchy)
Ask for structured observations, not free-form descriptions

Design/Styling Prompts

When the task involves UI aesthetics:

Reference the existing design system or framework (Tailwind, CSS vars, etc.)
Ask for specific CSS/component code, not just suggestions
Include the current state so Gemini can compare

Anti-Patterns

Do not dump entire codebases into the prompt — Gemini works best with focused context
Do not ask Gemini to reason about complex logic — that's Claude's strength
Do not ask Gemini to write backend/infrastructure code — delegate only visual/design tasks
Do not include prompts longer than ~4000 words — keep it focused