Images & thumbnails: agent skills for AI images and YouTube thumbnails

For AI agents: find the right skill in one call — curl -s "https://varg.ai/how-to/api/search?q=<what the person wants>" returns the matching how-to page and the best skills with install commands, as markdown (Russian works too). Full index: https://varg.ai/how-to/llms.txt. Every page has a markdown twin: add .md to its path. Before generating anything, show the person the options on the how-to page.

Image models, YouTube thumbnails, covers, carousels, upscaling and background removal. 50 skills for Claude Code, Codex and Grok, by type: workflows take a brief to a finished video, effects and styles change how it looks, prompts and guides teach the agent the craft, models run the generation.

How-to tutorials

• How to make YouTube video thumbnails with AI: High-CTR thumbnails and vertical covers, in a few options to choose from.
• How to make TikTok photo carousels with AI: Swipeable photo slideshows with hook text — the format app founders use for organic growth.

Skills (50)

Prompts (4)

• Replicate Prompt Images — Replicate: Write better image prompts and edits: photographic language, lighting, text rendering, consistent characters.
• FLUX Image Best Practices — Black Forest Labs: Prompt FLUX image models well: structure, JSON prompts, hex colors, typography and multi-reference edits.
• Douyin Cover Builder — 0x00000003, beta: Writes image-model prompts for Douyin covers: bold title, highlighted keywords, your face bottom right
• Composition & Framing — varg, beta: 34 composition cards — thirds, symmetry, leading lines, forced perspective, 2.39:1 to IMAX — for stills and video.

Styles (3)

• Canvas Design — Anthropic: Design posters and visual art as PNG or PDF from a written design philosophy, with 54 bundled fonts.
• GPT Image 2 Style Library — Canghe: Pick styles and industrial prompt templates for GPT Image 2/2.5 from a 530+ case library (a 35K-star repo).
• Anatomy Isometric Figures — Ryan: Detailed interactive isometric SVG figures that explain a technical idea as a physical object, with optional WebGL.

Guides (8)

• HyperFrames Media Use — HeyGen: Find or make every asset a video needs: music, SFX, voiceover, images, logos, avatars, transcripts, grades.
• PixiJS GIFs — PixiJS: Load, play and control animated GIFs inside a PixiJS v8 scene.
• App Store Screenshots — inference.sh: Make App Store and Google Play screenshots to exact specs: sizes, gallery order, mockups and previews.
• Book Cover Design — inference.sh: Design book covers that fit their genre: typography rules, sizing, thumbnail tests and AI iterations.
• Logo Design Guide — inference.sh: Design logos with AI image models: logo types, prompting techniques, scalability rules and iteration.
• OG Image Design — inference.sh: Design Open Graph and social share images to platform specs, with text placement and branding.
• YouTube Thumbnail Design — inference.sh: Design YouTube thumbnails that read on a phone: dimensions, contrast, safe zones, text and faces.
• Social Media Carousel — inference.sh: Design swipeable carousels for Instagram, LinkedIn and X: slide structure, hooks, text hierarchy.

Workflows (6)

• Higgsfield Brandkit — Higgsfield: Build a complete visual identity: palette, SVG logo marks, type, mockups, packaging and an editable brandbook.
• Higgsfield YouTube Thumbnail — Higgsfield: Design truthful, high-CTR YouTube thumbnails and vertical covers with up to three consistent faces.
• Slack GIF Creator — Anthropic: Make animated GIFs that pass Slack's limits: 128×128 emoji or 480×480 message GIFs, validated on export.
• ASO App Store Screenshots — Adam Lyttle: App Store screenshots from your codebase: finds the core benefits, pairs simulator shots, then composes and enhances.
• E-commerce Image Studio — buluslan: 39 e-commerce image templates (main images, lifestyle, A+, sizes, GIFs) with one style lock across a campaign set.
• Baoyu Cover Image — Jim Liu: Article cover images built from five dimensions (type, palette, rendering, text, mood) in 2.35:1, 16:9 or 1:1.

Models & tools (29)

• varg AI — varg: The official varg skill: videos, images, speech and music through one API key, rendered in the cloud or locally.
• Higgsfield Generate — Higgsfield: Generate images, video, 3D and audio across 30+ models, plus ad videos and virality scoring, from one CLI.
• fal Models Catalog — fal: Choose the right fal endpoint for any modality from curated picks: image, video, 3D, audio and vision.
• fal genmedia — fal: Run any of 1,200+ fal endpoints from the terminal: search, inspect schema, price, run, poll and download.
• fal Model Routing — fal: Pick default fal endpoints per job: premium stills, fast drafts, edits, video, lip-sync and campaign work.
• Gemini API Dev — Google Gemini: Write current Gemini API code, including Nano Banana image generation, Omni video and Gemini TTS.
• Melies — Melies: Generate images, video, movie posters and thumbnails from the terminal with 148 AI actors and style flags.
• AI Image Generation — inference.sh: Generate images with 50+ models from one CLI: GPT Image 2.5, FLUX, Gemini, Grok, Seedream, Reve and more.
• Background Removal — inference.sh: Cut subjects out of photos into clean transparent PNGs with BiRefNet.
• FLUX Image — inference.sh: Generate and restyle images with FLUX Dev LoRA and FLUX.2 Klein, including custom LoRA styles.
• GPT Image — inference.sh: Generate and edit images with OpenAI GPT Image 2.5 and GPT Image 2 from the command line.
• Image Upscaling — inference.sh: Upscale and enhance low-resolution images with Real-ESRGAN, Thera, Topaz or the FLUX upscaler.
• Nano Banana — inference.sh: Generate and edit images with Google's Gemini 3 Pro Image and Gemini 2.5 Flash Image, multi-image input too.
• Nano Banana 2 — inference.sh: Generate and edit images with Nano Banana 2, with up to 14 reference images and search grounding.
• P-Image — inference.sh: Generate and edit images fast with Pruna's P-Image models, including LoRA styles and multi-image edits.
• Qwen Image 2 — inference.sh: Generate and edit images with Alibaba Qwen-Image 2.0: the fast model, or Pro for cleaner text rendering.
• Qwen Image 2 Pro — inference.sh: Make posters, banners and text-heavy designs with Qwen-Image 2.0 Pro's precise text rendering.
• inference.sh CLI — inference.sh: Log in to inference.sh and run its catalog of AI apps — image, video, audio, LLMs, search — from one CLI.
• Runway Dev — Runway: Build Runway's image, video and audio models into your app: live API contracts, SDK and server-side keys.
• Runway Model Routers — Runway: Send generation through a saved Runway Model Router config instead of hard-coding one model.
• Runway Dev Models — Runway: Add a direct Runway model endpoint to your app for image, video or audio, following the live contract.
• Runway Recipes — Runway: Integrate Runway's prebuilt multi-step image and video Recipes using the live Recipe guide.
• Runway Media — Runway: Generate and edit images, video, music, speech and sound effects from chat with Runway's MCP tools.
• Runway Studio Skills — Runway: Run Runway's API from Python scripts: text, image and video-to-video, character performance, images, TTS.
• BFL API — Black Forest Labs: Integrate the FLUX API correctly: endpoints, async polling, webhooks, rate limits, errors and regions.
• Krea Generate — Krea: Pick the right Krea model and prompt for any image, video, edit or upscale before you generate.
• Kling AI — Kling AI, beta: Official Kling CLI: text/image-to-video, Omni reference and edit, images, reusable subjects, quota
• dLazy Jimeng Text-to-Image — dlazyai, beta: Generates images with Jimeng (Dreamina) from a prompt and up to 10 reference images, through the dLazy CLI
• dLazy Kling Image O1 — dlazyai, beta: Kling o1 text-to-image and reference-image editing at 2K via the dLazy CLI