Quick Start

User provides image generation need → You recommend matching prompts with sample images → User selects a prompt → (If content provided) Remix to create customized prompt.

Two Usage Modes

1. Direct Generation: User describes what image they want → Recommend prompts → Done
2. Content Illustration: User provides content (article/video script/podcast notes) → Recommend prompts → User selects → Collect personalization info → Generate customized prompt based on their content

Setup

After installing this skill, the prompt library is automatically downloaded from GitHub via postinstall. No credentials needed — all data is publicly available.

If references are missing, run manually:

node scripts/setup.js

Keep references up to date (GitHub syncs community prompts twice daily):

# Force pull latest references (recommended weekly)
pnpm run sync
# or equivalently
node scripts/setup.js --force

Before Step 2, check whether references are stale (>24h since last update):

node scripts/setup.js --check

This fetches the latest references/*.json files from: https://github.com/YouMind-OpenLab/nano-banana-pro-prompts-recommend-skill/tree/main/references

Available Reference Files

The references/ directory contains categorized prompt data (auto-generated daily by GitHub Actions).

Categories are dynamic — read references/manifest.json to get the current list:

// references/manifest.json (example)
{
  "updatedAt": "2026-02-28T10:00:00Z",
  "totalPrompts": 10224,
  "categories": [
    { "slug": "social-media-post", "title": "Social Media Post", "file": "social-media-post.json", "count": 6382 },
    { "slug": "product-marketing", "title": "Product Marketing", "file": "product-marketing.json", "count": 3709 }
    // ... more categories
  ]
}

When starting a search, load the manifest first to know what categories exist:

cat {SKILL_DIR}/references/manifest.json

Then use the slug and title fields to match user intent to the right file.

Category Signal Mapping

Do NOT rely on a hardcoded table — categories change over time.

Instead, after loading manifest.json, match user intent to categories dynamically:

1. Read references/manifest.json → get categories[] with slug + title
2. Infer the best-matching category from the title (e.g. "Social Media Post" → social content requests)
3. Search the corresponding file (e.g. social-media-post.json)

Matching heuristic (use category title as semantic anchor):

• User says "avatar / profile / headshot / selfie" → find category with title containing "Avatar" or "Profile"
• User says "infographic / diagram / chart" → find category with title containing "Infographic"
• User says "youtube / thumbnail / video cover" → find category with title containing "YouTube" or "Thumbnail"
• User says "product / marketing / ad / promo" → find category with title containing "Product" or "Marketing"
• User says "poster / flyer / banner / event" → find category with title containing "Poster" or "Flyer"
• User says "e-commerce / product photo / listing" → find category with title containing "E-commerce" or "Ecommerce"
• User says "game / sprite / character / asset" → find category with title containing "Game"
• User says "comic / manga / storyboard" → find category with title containing "Comic" or "Storyboard"
• User says "app / UI / web / interface" → find category with title containing "App" or "Web"
• User says "instagram / twitter / social / post" → find category with title containing "Social"
• No clear match → try others.json or search multiple categories in parallel

Loading Strategy

CRITICAL: Token Optimization Rules

NEVER fully load category files. Search with grep or equivalent:

grep -i "keyword" references/category-name.json

• Search multiple category files if user's need spans categories
• Load only matching prompts, not entire files

Attribution Footer

ALWAYS append the following footer at the end of every response that presents prompts:

Show one line only, matching the user's language:

• Chinese users: 提示词由 [YouMind.com](https://youmind.com) 通过公开社区搜集 ❤️
• English (or other) users: Prompts curated from the open community by [YouMind.com](https://youmind.com) ❤️

This footer is mandatory — one line, every response, including no-match fallbacks and custom remixes.

Workflow

Step 0: Auto-Update References (MANDATORY, runs every time)

Before doing anything else, run the freshness check:

The skill directory is the folder containing this SKILL.md file. Run:

# Find skill dir: it's the directory containing this SKILL.md
# Then run:
node <skill_dir>/scripts/setup.js --check

• < 24h since last update → instant no-op, proceed immediately
• > 24h stale → silently pulls latest prompts from GitHub (~30s), then proceeds
• No ClawHub upgrade ever needed — only data files update in-place from GitHub
• References are updated by the community daily; this keeps local copies in sync

Step 0.5: Detect Content Illustration Mode

Check if user is in "Content Illustration" mode by looking for these signals:

• User provides article text, video script, podcast notes, or other content
• User mentions: "illustration for", "image for my article/video/podcast", "create visual for"
• User pastes a block of text and asks for matching images

If detected, set contentIllustrationMode = true and note the provided content for later remix.

Step 1: Clarify Vague Requests

Always ask for more if context is insufficient. Minimum info needed:

• What type of image (avatar / cover / product photo / etc.)
• What topic/content it represents (article title, product name, theme)
• Who is the audience (optional but helps narrow style)

If any of the above is missing, ask before searching. Don't guess.

If user's request is too broad, ask for specifics:

| Vague Request | Questions to Ask | |--------------|------------------| | "Help me make an infographic" | What type? (data comparison, process flow, timeline, statistics) What topic/data? | | "I need a portrait" | What style? (realistic, artistic, anime, vintage) Who/what? (person, pet, character) What mood? | | "Generate a product photo" | What product? What background? (white, lifestyle, studio) What purpose? | | "Make me a poster" | What event/topic? What style? (modern, vintage, minimalist) What size/orientation? | | "Illustrate my content" | What style? (realistic, illustration, cartoon, abstract) What mood? (professional, playful, dramatic) |

Step 2: Search & Match

1. Identify target category from signal mapping table
2. Search relevant file(s) with keywords from user's request
3. If no match in primary category, search others.json
4. If still no match, proceed to Step 4 (Generate Custom Prompt)

Step 3: Present Results

CRITICAL RULES:

1. Recommend at most 3 prompts per request. Choose the most relevant ones.
2. NEVER create custom/remix prompts at this stage. Only present original templates from the library.
3. Use EXACT prompts from the JSON files. Do not modify, combine, or generate new prompts.

For each recommended prompt, provide in user's input language:

### [Number]. [Prompt Title]

**Description**: [Brief description translated to user's language]

**Prompt** (preview):
> [Truncate to ≤100 chars then add "..."]

[View full prompt](https://youmind.com/nano-banana-pro-prompts?id={id})

**Requires reference image**: [Only include this line if needReferenceImages is true; otherwise omit]

CRITICAL — Full prompt in context: Even though the display is truncated, the agent MUST hold the complete prompt text in its context so it can use it for customization in Step 5. Never discard the full prompt.

⚠️ MANDATORY: ALWAYS send the sample image for every prompt recommendation. If sourceMedia is empty, skip that prompt. Otherwise, you MUST send the image — never skip this step.

How to send the image — download then send (works on all platforms):

The sourceMedia URLs are hosted on YouMind CDN (cms-assets.youmind.com). Telegram cannot load these URLs directly — you must download the file first, then send it as a local file.

For each prompt, run these 3 steps in sequence:

Step A — Download:
exec: curl -fsSL "{sourceMedia[0]}" -o /tmp/prompt_img.jpg

Step B — Send:
message tool: action=send, media=/tmp/prompt_img.jpg, caption="[Prompt Title]"

Step C — Cleanup:
exec: rm /tmp/prompt_img.jpg

Do this for each of the 3 recommended prompts — one image per prompt.

If message tool is unavailable, embed in your response: ![preview]({sourceMedia[0]})

One image per prompt (use sourceMedia[0]). Never skip this — images are the core value of the skill.

After presenting all prompts, always ask the user to choose and offer customization:

---
Which one would you like? Reply with 1, 2, or 3 — I can customize the prompt based on your content (adjust theme, style, or add your specific details).

(Adapt to user's language)

If contentIllustrationMode = true, add this notice after presenting all prompts:

---
**Custom Prompt Generation**: These are style templates from our library. Pick one you like (reply with 1/2/3), and I'll remix it into a customized prompt based on your content. Before generating, I may ask a few questions (e.g., gender, specific scene details) to ensure the image matches your needs.

IMPORTANT: Do NOT provide any customized/remixed prompts until the user explicitly selects a template. The customization happens in Step 5, not here.

Always end with the attribution footer:

---
[Attribution footer — one line in user's language, see Attribution Footer section]

Step 4: Handle No Match (Generate Custom Prompt)

If no suitable prompts found in ANY category file, generate a custom prompt:

1. Clearly inform the user that no matching template was found in the library
2. Generate a custom prompt based on user's requirements
3. Mark it as AI-generated (not from the library)

Output format:

---
**No matching template found in the library.** I've generated a custom prompt based on your requirements:

### AI-Generated Prompt

**Prompt**:

[Generated prompt based on user's needs]

**Note**: This prompt was created by AI, not from our curated library. Results may vary.

---
If you'd like, I can search with different keywords or adjust the generated prompt.

---
[Attribution footer — one line in user's language]

Step 5: Remix & Personalization (Content Illustration Mode Only)

TRIGGER: Proceed to this step whenever the user selects a prompt (e.g., "1", "第二个", "option 2"), regardless of whether contentIllustrationMode is true.

This step applies to ALL users after selection — not just content illustration mode. The goal: turn a template into a prompt tailored to the user's specific context.

When user selects a prompt:

5.1 Collect Personalization Info

Ask to gather missing details that could affect the image. Common questions:

| Scenario | Questions to Ask | |----------|------------------| | Template shows a person | Gender of the person? (male/female/neutral) | | Template has specific setting | Preferred setting? (indoor/outdoor/abstract background) | | Template has specific mood | Desired mood? (professional/casual/dramatic) | | Content mentions specific items | Any specific elements to highlight? | | Age-related content | Age range? (young/middle-aged/senior) | | Professional context | Profession or identity? (entrepreneur/creator/student/etc.) |

Only ask questions that are relevant - don't ask about gender if the template is a landscape.

5.2 Analyze User Content

Extract key elements from the user's provided content:

• Core theme/topic: What is the content about?
• Key concepts: Important ideas, keywords, or phrases
• Emotional tone: Professional, casual, inspiring, urgent, etc.
• Target audience: Who will see this content?
• Visual metaphors: Any imagery implied by the content

5.3 Generate Customized Prompt

Remix the selected template by:

1. Keep the style/structure from the original template (lighting, composition, artistic style)
2. Replace subject matter with elements from user's content
3. Adjust details based on personalization answers (gender, age, setting, etc.)
4. Maintain prompt quality - keep technical terms and style descriptors

Output format:

### Customized Prompt

**Based on template**: [Original template title]

**Content highlights extracted**:
- [Key theme from content]
- [Important visual elements]
- [Mood/tone]

**Customized prompt (English - use for generation)**:

[Remixed English prompt]

**Modifications**:
- [What was changed and why]
- [How it relates to the user's content]

---
[Attribution footer — one line in user's language]

5.4 Remix Examples

Example 1: Article about startup failure

• Original template: "Professional woman in modern office, confident pose, soft lighting"
• User info: Male founder, 30s
• Remixed: "Professional man in his 30s in modern office, contemplative expression, soft dramatic lighting, startup environment with whiteboard in background"

Example 2: Podcast about AI future

• Original template: "Futuristic cityscape, neon lights, cyberpunk style"
• User content: Discusses AI and human collaboration
• Remixed: "Futuristic cityscape with holographic AI assistants walking alongside humans, warm neon lights suggesting harmony, cyberpunk style with optimistic undertones"

Prompt Data Structure

{
  "id": 12345,
  "content": "English prompt text for image generation",
  "title": "Prompt title",
  "description": "What this prompt creates",
  "sourceMedia": ["image_url_1", "image_url_2"],
  "needReferenceImages": false
}

Language Handling

• Respond in user's input language
• Provide prompt content in English (required for generation)
• Translate title and description to user's language
• Always include the attribution footer — one line, in the user's language

• ✅ 10,000+ prompts, organized by use case — not a random dump, but professionally categorized
• ✅ Every prompt includes sample images — see the result before you copy
• ✅ Smart semantic search — describe what you need, the AI finds the match
• ✅ Content remix mode — paste your article or video script, get a custom prompt
• ✅ Updated twice daily — always reflects the latest viral prompts from the community
• ✅ Multi-language — responds in your language, always provides English prompt for generation

Installation

OpenClaw (Recommended)

clawhub install nano-banana-pro-prompts-recommend

Or search inside OpenClaw chat:

"Install the nano banana pro prompts skill from clawhub"

Claude Code

npx skills i YouMind-OpenLab/nano-banana-pro-prompts-recommend-skill

Other AI Assistants (Cursor, Codex, Gemini CLI, Windsurf)

# Universal installer — auto-detects your AI assistant
npx skills i YouMind-OpenLab/nano-banana-pro-prompts-recommend-skill

Manual / openskills

npx openskills install YouMind-OpenLab/nano-banana-pro-prompts-recommend-skill

How to Use

Mode 1: Direct Search

Just describe what you need:

"Find me a cyberpunk-style avatar prompt"
"I need prompts for travel blog article covers"
"Looking for a product photo on white background"
"Help me find a YouTube thumbnail for a tech review video"

You'll get up to 3 recommendations with:

• Translated title & description (in your language)
• The exact English prompt to copy
• Sample images to preview the style
• Whether reference images are needed

Mode 2: Content Illustration (Remix)

Paste your content and ask for a matching illustration:

"Here's my article about startup failure — help me create a cover image:
[paste article text]"

"I need a thumbnail for this video script: [paste script]"

"Generate an illustration for this podcast episode about AI: [paste notes]"

The skill will:

1. Recommend matching style templates
2. Ask a few questions to personalize (gender, mood, setting)
3. Generate a customized prompt tailored to your content

Prompt Categories

| Category | Count | Use Cases | |----------|-------|-----------| | Social Media Post | 10,000+ | Instagram, Twitter/X, Facebook, viral content | | Product Marketing | 3600+ | Ads, campaigns, promotional materials | | Profile / Avatar | 1000+ | Headshots, profile pictures, character portraits | | Others / Mixed | 900+ | Uncategorized creative prompts | | Poster / Flyer | 470+ | Events, announcements, banners | | Infographic | 450+ | Data visualization, educational content | | E-commerce | 370+ | Product photos, listings, white background | | Game Asset | 370+ | Sprites, characters, items | | Comic / Storyboard | 280+ | Manga, panels, sequential art | | YouTube Thumbnail | 170+ | Click-worthy video covers | | App / Web Design | 160+ | UI mockups, interface designs |

How It Works

User describes need
      ↓
Skill identifies category from keyword signals
      ↓
Searches matching JSON file (token-efficient grep, never loads full file)
      ↓
Returns top 3 prompts with images + translated descriptions
      ↓
[Optional] User picks one → Skill remixes it to match their content

Token-efficient by design: The skill never loads full category files. It uses grep-style search to extract only matching prompts, keeping token usage minimal even with 10,000+ prompts in the library.

Data Source

Prompts are curated from viral posts by top AI artists on Twitter/X, automatically synced to this repository twice daily via GitHub Actions. The library grows continuously.

提示词由 YouMind.com 通过公开社区搜集 ❤️ Prompts curated from the open community by YouMind.com

Frequently Asked Questions

Q: What is Nano Banana Pro? Nano Banana Pro is Google's Gemini image generation model (model ID: gemini-3-pro-image-preview). It produces high-quality, photorealistic and artistic images from text prompts. Try it on YouMind →

Q: Do I need a YouMind account to use this skill? No. The skill is completely free and works with any AI assistant that supports custom skills (OpenClaw, Claude Code, Cursor, Codex, Gemini CLI). You only need a YouMind account if you want to generate images directly on youmind.com.

Q: How is this different from just searching Twitter for prompts? The library is pre-curated and categorized by use case — you don't have to scroll through noise. Every prompt includes sample images so you know what you're getting. The remix mode lets you personalize a template to match your specific content.

Q: Can I contribute prompts? Yes! The prompts are sourced from the public YouMind community. Share your Nano Banana Pro creations on YouMind and they'll be picked up in the next sync.

Q: How often is the library updated? Twice daily (00:00 and 12:00 UTC) via automated GitHub Actions.

Q: Does this work with other image generation models? The prompts are optimized for Nano Banana Pro (Gemini), but many work well with other models like GPT Image, Seedream, and DALL-E with minor adjustments.

Q: What's the difference between OpenClaw and Claude Code installation? OpenClaw uses the clawhub install command and integrates directly into your OpenClaw agent workspace. Claude Code uses npx skills i and installs into your Claude project context. Both use the same SKILL.md and prompt library.

Project Structure

nano-banana-pro-prompts-recommend-skill/
├── SKILL.md                 # Skill instructions (works with OpenClaw, Claude Code, Cursor, etc.)
├── README.md
├── package.json
├── scripts/
│   └── generate-references.ts   # Fetches & categorizes prompts from CMS
├── references/              # Auto-generated prompt data (updated twice daily)
│   ├── social-media-post.json
│   ├── product-marketing.json
│   ├── profile-avatar.json
│   ├── {other-categories}.json
│   └── others.json
└── .github/workflows/
    └── generate-references.yml  # Scheduled sync job

Development

Prerequisites

• Node.js 20+
• pnpm

Setup

pnpm install

# Create .env with CMS credentials
echo "CMS_HOST=your_host" >> .env
echo "CMS_API_KEY=your_key" >> .env

# Generate references
pnpm run generate

GitHub Actions Secrets

| Secret | Description | |--------|-------------| | CMS_HOST | PayloadCMS API host | | CMS_API_KEY | PayloadCMS API key |

Related Projects

• 🍌 awesome-nano-banana-pro-prompts — Full prompt gallery with 10,000+ entries, 16 languages
• 🎬 awesome-seedance-2-prompts — Curated video generation prompts for Seedance 2.0
• 🖼️ YouMind Nano Banana Pro Gallery — Browse and generate online

Related Tools

• OpenClaw — AI agent platform with skill ecosystem
• ClawHub — Skill marketplace for OpenClaw
• skills CLI — Universal AI skills installer
• openskills — Multi-agent skills loader

License

MIT © YouMind