Youtube Thumbnail Design

YouTube thumbnail design with specific dimensions, contrast rules, and mobile preview optimization. Covers safe zones, text placement, face expression psycho...

Security Scan

The skill's stated purpose (AI-assisted YouTube thumbnail design) matches the instructions, but the runtime install pattern (curl | sh from a remote host) and use of a third-party CLI that may request logins are disproportionate and warrant caution.

Purpose & Capability

Name and description match the instructions: the SKILL.md solely describes generating thumbnails via the inference.sh (infsh) CLI and provides composition/visual guidance; no unrelated credentials, binaries, or system access are requested.

Instruction Scope

Instructions focus on thumbnail generation (prompts, sizes, composition rules, sample infsh commands). They do not ask the agent to read arbitrary host files, environment variables, or send data to unexpected endpoints in the visible content. (Note: SKILL.md was truncated — additional instructions past the truncation could change this assessment.)

Install Mechanism

The Quick Start recommends piping a remote install script (curl -fsSL https://cli.inference.sh | sh). That pattern executes remote code without local checksum verification and is high-risk. The file claims a checksum is available on dist.inference.sh, but the provided command bypasses any verification step. The install source is not a well-known package manager; this is proportionally excessive for a thumbnail design instruction-only skill.

Credentials

The skill declares no required env vars, credentials, or config paths. However, it invokes infsh login which implies the CLI may request or create credentials/tokens at runtime; that is expected for a model-inference service but users should be aware the CLI may obtain/store auth tokens.

Persistence & Privilege

always is false (good). disable-model-invocation is false (default), so the agent could invoke this skill autonomously when eligible — that is normal but increases blast radius if combined with the risky install pattern or if the CLI stores tokens. The skill does not request system-wide config changes in the visible content.

What to consider before installing

This skill appears to do what it says (generate thumbnail images and give design guidance) but take these precautions before installing or running it: - Avoid running the recommended curl | sh pipeline. That pattern executes remote code immediately and bypasses verification. Instead, manually download the CLI binary or follow the manual install steps and verify the SHA-256 checksums on the project's dist.inference.sh site before running any installer. - Expect the infsh CLI to require login/auth tokens. If you must authenticate, prefer short-lived credentials, inspect where tokens are stored, or run the CLI in an isolated environment (container or VM) so stored tokens cannot be reused by other tools. - Review the remote domain (inference.sh / dist.inference.sh) and, if possible, inspect the CLI source or package artifacts before running. If you cannot verify provenance, run the workflow in a sandbox. - Because part of the SKILL.md was truncated, consider asking the publisher for the full instructions and confirm there are no additional steps that read local files or post content to unexpected endpoints. Given the insecure default install guidance and potential for the CLI to obtain credentials, proceed with caution — the skill is coherent but the install/runtime recommendations are risky unless you verify them manually.

Create high-CTR YouTube thumbnails with AI image generation via inference.sh CLI.

Quick Start

curl -fsSL https://cli.inference.sh | sh && infsh login

# Generate a thumbnail
infsh app run falai/flux-dev-lora --input '{
  "prompt": "YouTube thumbnail style, close-up of a person with surprised excited expression looking at a glowing laptop screen, vibrant blue and orange color scheme, dramatic studio lighting, shallow depth of field, high contrast, cinematic",
  "width": 1280,
  "height": 720
}'

Install note: The install script only detects your OS/architecture, downloads the matching binary from dist.inference.sh, and verifies its SHA-256 checksum. No elevated permissions or background processes. Manual install & verification available.

Specifications

Spec	Value
Dimensions	1280 x 720 px (minimum)
Recommended	1920 x 1080 px
Aspect ratio	16:9
Max file size	2 MB
Formats	JPG, GIF, PNG

The 120px Test

Your thumbnail appears at roughly 120px wide on mobile — that's how most viewers first see it.

At 120px, viewers must be able to identify:

1. The mood/emotion (from colors and expression)
2. The general subject (from composition)
3. The text (if any — only if large enough)

Test: view your thumbnail at 120px width. If it's a muddy blur, redesign.

Safe Zones

┌─────────────────────────────────────────────┐
│                                             │
│   ✅ SAFE FOR TEXT AND KEY ELEMENTS         │
│                                             │
│                                             │
│                                             │
│                                             │
│                                       ┌───┐ │
│                                       │ ⏱ │ │ ← Timestamp overlay
│                              ┌────────┴───┘ │    (bottom-right)
│   ┌────┐                     │  DURATION    │
│   │ CH │ Chapter marker      └──────────────│
└───┴────┴────────────────────────────────────┘
     ↑ Bottom-left: chapter/progress markers

Avoid placing critical elements in:

• Bottom-right corner (video duration timestamp)
• Bottom-left corner (chapter markers, progress bar)
• Extreme edges (cropping varies by device)

Color Strategy

High-Contrast Pairs That Work

Combination	Mood	Best For
Yellow + Black	Urgency, attention	Tech, business, lists
Red + White	Energy, excitement	Entertainment, reactions
Blue + Orange	Professional contrast	Education, tutorials
Green + White	Growth, money	Finance, success stories
Purple + Yellow	Premium, creative	Design, art, creativity
White + Dark	Clean, minimal	Luxury, minimalist channels

Color Rules

• Background and text/subject should be complementary or high-contrast
• Avoid same-temperature colors touching (red on orange = mud)
• Use 3 colors maximum per thumbnail
• Saturate more than real life — thumbnails compete with bright UI

Text on Thumbnails

When to Use Text

• Lists/numbers: "7 Tips", "Top 10"
• Strong opinions: "STOP Doing This"
• Results: "$10K in 30 Days"
• Comparisons: "vs" between two things

When NOT to Use Text

• The video title already says it (redundant)
• The emotion/visual tells the story
• You can't make it large enough to read at 120px

Text Rules

Rule	Reason
Max 6 words	Readability at thumbnail size
Min 60pt equivalent	Must be legible at 120px width
Bold sans-serif font	Thin fonts disappear at small sizes
Contrast stroke/shadow	Ensures readability on any background
No small text	If it's not readable small, cut it

Face Expression Psychology

Thumbnails with faces get higher CTR than faceless thumbnails. Expression matters:

Expression	CTR Impact	Best For
Surprise/shock	Highest	Reaction, reveal, discovery content
Curiosity	High	Tutorial, how-to, tips
Excitement	High	Unboxing, reviews, announcements
Concern/worry	Medium-high	Warning, mistake, problem content
Confidence	Medium	Expert advice, authority content
Neutral	Lowest	Avoid unless your brand is minimalist

Face Composition Rules

• Face should fill 30-50% of the thumbnail
• Eyes looking toward the text or subject (directs viewer attention)
• Eyes looking at camera = connection. Eyes looking at object = curiosity.
• Place face on one side (usually left), text or subject on the other

# Generate a face-forward thumbnail
infsh app run falai/flux-dev-lora --input '{
  "prompt": "close-up portrait of a man with genuinely surprised expression, mouth slightly open, raised eyebrows, looking at camera, left side of frame, vibrant teal background, dramatic rim lighting, YouTube thumbnail style, high contrast, cinematic",
  "width": 1280,
  "height": 720
}'

# Generate a face-looking-at-subject thumbnail
infsh app run bytedance/seedream-4-5 --input '{
  "prompt": "person looking amazed at a glowing holographic chart showing upward growth, dramatic blue and green lighting, right side profile view, dark background, tech aesthetic, high energy",
  "size": "2K"
}'

Thumbnail Patterns by Content Type

Tutorial / How-To

infsh app run falai/flux-dev-lora --input '{
  "prompt": "overhead flat lay of organized workspace with laptop showing code editor, colorful sticky notes, coffee cup, clean bright background, professional setup, tutorial style composition, warm lighting",
  "width": 1280,
  "height": 720
}'

Before/After

infsh app run falai/flux-dev-lora --input '{
  "prompt": "split composition, left side dark and messy disorganized desk, right side bright clean organized minimalist workspace, dramatic contrast between chaos and order, clear dividing line in center, high contrast",
  "width": 1280,
  "height": 720
}'

Product Review / Comparison

infsh app run falai/flux-dev-lora --input '{
  "prompt": "two products facing each other with dramatic lighting and sparks between them, competition battle concept, dark background with colorful rim lighting, versus comparison style, high energy, product photography",
  "width": 1280,
  "height": 720
}'

Listicle / Number

infsh app run falai/flux-dev-lora --input '{
  "prompt": "dynamic arrangement of 7 different colorful objects floating in space against dark gradient background, each item distinct and clearly separated, energetic composition, vibrant saturated colors, studio lighting",
  "width": 1280,
  "height": 720
}'

A/B Testing

Test one variable at a time:

Variable	Test A vs B
Face vs No face	Same composition, with/without person
Expression	Surprise vs curiosity
Color scheme	Warm vs cool palette
Text vs No text	With/without text overlay
Background	Bright vs dark
Composition	Left-facing vs right-facing subject

# Generate variant A
infsh app run falai/flux-dev-lora --input '{
  "prompt": "..., bright yellow background, ...",
  "width": 1280, "height": 720
}' --no-wait

# Generate variant B (same prompt, different background)
infsh app run falai/flux-dev-lora --input '{
  "prompt": "..., dark navy background, ...",
  "width": 1280, "height": 720
}' --no-wait

Thumbnail Checklist

• 1280x720 minimum (1920x1080 preferred)
• Under 2MB file size
• Passes the 120px squint test
• No critical elements in bottom-right (timestamp) or bottom-left (chapter)
• Max 3 colors, high contrast
• Text (if any) is max 6 words, bold, with contrast stroke
• Face expression matches content energy (if applicable)
• Doesn't duplicate the video title
• Stands out from surrounding thumbnails (check your niche)
• Works on both light and dark YouTube backgrounds

Common Mistakes

Mistake	Problem	Fix
Too much text	Unreadable at thumbnail size	Max 6 words or no text
Low contrast	Disappears in the feed	Use complementary colors
Cluttered composition	Eye doesn't know where to look	One focal point
Generic stock photo feel	No personality, gets skipped	Authentic expressions, unique angles
Tiny details	Lost at 120px	Bold, simple shapes
Same style every video	Viewer fatigue	Vary within brand guidelines
Misleading thumbnail	Kills trust, hurts retention	Match the actual content

Related Skills

npx skills add inference-sh/skills@ai-image-generation
npx skills add inference-sh/skills@image-upscaling
npx skills add inference-sh/skills@prompt-engineering

Browse all apps: infsh app list