smixs

smixs / visual-skills

Public

Professional Claude Skills for AI image and video prompting. Supports Nano Banana (Gemini 3 Pro/Flash), GPT Image 2, Seedance 2.0, Kling 3.0 (multi-shot + native dialogue), Veo. Works with Claude Code, Cursor, Windsurf, OpenCode, Hermes-agent.

github.comsmixsvisual-skills ai-image-generation ai-prompts ai-video-generation claude claude-agent-sdk
12
2
100% credibility
Found Apr 27, 2026 at 12 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
AI Summary

This repository offers markdown-based skills for Claude AI to generate production-ready prompts for leading AI image and video generation models.

How It Works

1
🔍 Discover Visual Skills

While looking for ways to make stunning images and videos with AI, you stumble upon this handy set of tools for your AI assistant.

2
📥 Download the Tools

Grab the simple packages for image creation or video making right from the collection.

3
Pick Image or Video
🖼️
Go for Images

Select the image tool to craft prompts for photos, designs, posters, or edits.

🎬
Go for Videos

Select the video tool to build prompts for short clips, stories, or animations.

4
➕ Add to Your AI Helper

Drop the chosen tool into your AI assistant's collection so it's ready to use.

5
💭 Describe Your Idea

Chat with your AI about what you envision, like a product poster or a dramatic scene, and it thinks deeply to create the best description.

6
📋 Get Ready-to-Use Prompt

Receive a perfect, copy-paste description tailored for top image or video creators.

✨ Create Stunning Visuals

Paste it in and watch your ideas come alive as professional-quality images or videos.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 12 to 12 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is visual-skills?

Visual-skills delivers two Claude Skills that generate production-ready prompts for AI image and video models like Nano Banana, GPT Image 2, Seedance 2.0, Kling 3.0, and Veo. It solves sloppy prompting by auto-selecting the best model, applying exact syntax (like Kling's multi-shot dialogue or Veo's JSON), and outputting copy-paste prompts—ditching vague "cinematic 4K" filler for precise visual skills appraisal. Built as markdown-based skills compatible with Claude Code, Cursor, Windsurf, and agents like Hermes, it works on any Claude professional account, respecting claude professional limits and usage plans.

Why is it gaining traction?

It stands out by enforcing model-specific rules—like Nano Banana for grounded scenes or Seedance for fast multi-shots—plus video dramaturgy with Murch's Rule of Six, ensuring consistent character lock and no banned fluff phrases. Developers notice instant gains in output quality for visual skills activities, without prompt trial-and-error, and seamless installs via .skill packages or folder drops. The professional GitHub readme and profile setup make it easy to adapt for team workflows.

Who should use this?

AI content teams, creative directors, and copywriters building ad creatives, product mockups, or multi-clip stories. Frontend devs in Cursor or Windsurf prototyping UI visuals or storyboards. Marketers on claude professional plan needing efficient visual skills assessment for campaigns, hitting claude professional usage limits less often with optimized prompts.

Verdict

Worth a quick install for Claude users doing visuals—solid docs and MIT license make forking easy, despite 12 stars and 1.0% credibility signaling early maturity. Test on low-stakes claude professional plan cost jobs first; it punches above its weight for prompt pros.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.