Headlines & callouts prompts for AI video
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.mdto its path. Before generating anything, show the person the options on the how-to page. From a terminal:curl -fsSL https://varg.ai/install | shinstalls the vargai CLI (https://varg.ai/how-to/cli.md):vargai searchandvargai install <slug>with no key, every varg model aftervargai login.
6 viral prompts with results and 0 prompt skills for headlines & callouts: words over footage: headlines, key phrases, lower thirds, highlights and callouts that point at the thing that matters.
Viral headlines & callouts prompts with results (6)
Full prompts, verbatim from their authors, each with the model it ran on and the post's own result, best first (likes plus three times saves). Credit the author when you reuse one. Examples to remake, with varg's remake prompts, are on Headlines & callouts.
Photo-to-poster split: subject crosses a geometric window
- Model: GPT Image 2 · image · photo plus minimalist poster split (3:4)
- Run on varg:
gpt_image_2_5(toolimage, aspect 3:4) — spec https://varg.ai/models/gpt_image_2_5.md - Post: @xiaoxiaodong, 2026-08-20 — 134,177 views, 540 likes, 805 bookmarks
- Who: xiaoxiaodong (30.9k followers), AI creator posting GPT Image prompts
- Found on: YouMind-OpenLab/awesome-gpt-image-2
- Result: https://pbs.twimg.com/media/HQKlZxlbwAAkQVH.jpg (+3 more in the post)
- Inputs: one photo per poster
- Sections: format, upper half rules, lower half concept, composition, palette, typography, overall mood
- Why it spread: 134.2k views and 805 bookmarks: top half keeps your photo, bottom half rebuilds the subject breaking out of a narrow pale color block, with a small English title.
请将我上传的每一张照片分别制作成一张独立的高级设计海报,不多图拼接,每张照片单独输出。整体采用 3:4 竖版构图,上下两个区域高度严格 1:1,各占画面 50%。
上半部分保留原始照片,保持主体结构、真实质感、自然光影和原有色彩氛围,仅进行轻微高级摄影调色,使其具有高端品牌广告、艺术杂志与展览视觉的精致质感。为适配画幅,可自然扩展天空、地面或环境背景,主体本身保持不变。
下半部分提取照片中最具识别性的**主体、轮廓、姿态与叙事关系**,重构为具有高级商业审美的极简视觉海报。核心语言是 **“真实物象穿越几何情绪窗口”**:在画面中设置一个窄长、克制、低饱和的几何色块,作为视觉锚点与空间容器。**色块方向不固定,可为横向、竖向或斜向**,根据主体的重心、动作方向、轮廓走势与版面节奏自然决定。主体以真实摄影感或精致写实方式进入色块之中,并局部越界、破框、伸出或穿过边界,形成自然生长感、空间穿透感与留白呼吸感。
构图保持极简、清晰、平衡,以几何色块、主体越界关系和大面积留白建立秩序。主体可居中、偏心、纵向生长或横向舒展,但始终保持单一焦点。背景以白色、浅灰或极浅暖色为主,可加入极轻微的文化纹样、线性图形、地形线、水波线、光影轮廓或抽象符号,作为若隐若现的辅助层次。
配色从上方照片中提取最有辨识度、最有精气神的颜色关系,转化为**浅色几何窗口 + 真实主体色彩 + 极少量中性色排版**的高级系统。色块颜色可柔化为雾蓝、浅青、米白、淡粉、暖灰、浅金或其他与原图气质一致的浅色调,整体轻盈、干净、现代、精致。
文字作为版式结构的一部分自然介入。可从照片的主题、情绪、动作、状态或象征意义中提炼一个简短英文标题,并搭配少量副标题、地点、状态词或微型注释。标题可横排、竖排或沿色块与主体关系排布,辅助信息以小字号、细字重、舒展字距规整排列,使其呈现高端地产、奢侈品、美学杂志或节气视觉般的高级编辑感。
整体气质强调**东方留白、现代秩序、自然生命力与高级商业美感**。无论主体是人物、动物、植物、建筑、器物、食物、交通工具或自然景观,都应保持真实摄影与几何情绪窗口之间清晰、优雅、可迁移的视觉呼应。
English: Make each photo I upload into its own separate high-end design poster; do not combine several images, output each photo separately. Use an overall 3:4 vertical composition, with the upper and lower areas strictly equal in height at 1:1, each taking 50% of the frame.
The upper half keeps the original photo, preserving the subject's structure, real texture, natural light and shadow and the original color mood, with only a light high-end photographic grade, giving it the refined feel of high-end brand advertising, art magazines and exhibition visuals. To fit the format, the sky, ground or surroundings may be extended naturally, while the subject itself stays unchanged.
The lower half extracts the photo's most recognizable **subject, outline, pose and narrative relationship** and rebuilds them into a minimalist visual poster with high-end commercial aesthetics. The core language is **"a real object passing through a geometric emotional window"**: place one narrow, long, restrained, low-saturation geometric color block in the frame as the visual anchor and spatial container. **The block's direction is not fixed; it can be horizontal, vertical or diagonal**, decided naturally by the subject's center of gravity, direction of movement, outline and the rhythm of the layout. The subject enters the color block with a real photographic feel or refined realism, and partly crosses its edge, breaks the frame, reaches out or passes through the boundary, creating a sense of natural growth, spatial penetration and breathing white space.
The composition stays minimal, clear and balanced, building order with the geometric color block, the subject's boundary-crossing and large areas of white space. The subject may be centered, off-center, growing vertically or stretching horizontally, but there is always a single focal point. The background is mainly white, light grey or a very pale warm tone, and may include very faint cultural patterns, linear graphics, contour lines, water-ripple lines, light-and-shadow outlines or abstract symbols as a barely-there supporting layer.
Colors are extracted from the photo above, taking its most recognizable and most spirited color relationships, and converted into a high-end system of **a pale geometric window + the subject's real colors + a very small amount of neutral-colored typography**. The color block may be softened into misty blue, pale cyan, off-white, light pink, warm grey, pale gold or another light tone that matches the original photo's mood; the whole is light, clean, modern and refined.
Text enters naturally as part of the layout structure. Distill a short English title from the photo's theme, emotion, action, state or symbolic meaning, and pair it with a little subtitle text, a place, a state word or micro-notes. The title may run horizontally, vertically or along the relationship between the color block and the subject; supporting information is set neatly in small sizes, light weights and open letter-spacing, giving it the high-end editorial feel of luxury real-estate, luxury-goods, aesthetics-magazine or seasonal-solar-term visuals.
The overall mood emphasizes **Eastern white space, modern order, natural vitality and high-end commercial beauty**. Whether the subject is a person, animal, plant, building, object, food, vehicle or natural landscape, keep a clear, elegant, transferable visual echo between the real photograph and the geometric emotional window.
Your life theme song as a CD jacket
- Model: ChatGPT with memory · image · fictional CD jacket from chat history
- Run on varg:
gpt_image_2_5(toolimage) — spec https://varg.ai/models/gpt_image_2_5.md - Post: @o8ee_L, 2026-09-22 — 61,794 views, 498 likes, 452 bookmarks
- Who: o8ee_L (2.4k followers), Japanese AI artist
- Result: https://pbs.twimg.com/media/HSyvPxVbgAA9KJo.jpg (+3 more in the post)
- Inputs: none (uses ChatGPT memory)
- Sections: Task from past chats, one flyer with title, artist and lyric, no-figures rule, mood and paired contrasts, design-not-illustration note, Japanese 2D graphic style, negative list, final quality goal.
- Why it spread: 62k views and 452 bookmarks: ChatGPT invents a song, artist name and tracklist from what it knows about you, then designs the CD.
Create a fictional life theme song and its CD jacket inspired by everything you know about me from our previous conversations. Imagine me as an original music artist and express my personal world, emotions, aesthetics, struggles, creativity, and sense of identity through an abstract visual concept.
Create ONE promotional flyer for the fictional song/CD. Include the fictional song title, artist name, and ONE short original lyric phrase that feels like a defining line from the song.
Do NOT include any characters, people, portraits, silhouettes, or human figures. Express the artist's world entirely through abstract visual elements, symbolic objects, typography, shapes, light, color, atmosphere, and composition.
The artwork should feel deeply personal, slightly surreal, dreamy, delicate, mysterious, and emotionally layered, combining cute, gothic, nostalgic, melancholic, and modern digital aesthetics. Reflect a contrast between fragility and strength, loneliness and connection, darkness and small traces of hope, while maintaining an elegant and cohesive visual identity.
Design it as a sophisticated fictional CD jacket / music promotional flyer rather than a literal illustration. Make the typography an important part of the composition. The lyric phrase should be visually integrated into the design, not treated like a large block of text.
Use a refined Japanese 2D graphic-art sensibility with clean shapes, crisp edges, sophisticated color design, subtle visual symbolism, and carefully balanced negative space.
Minimal texture. No grain, no photographic realism, no excessive detail, no 3D rendering, no glossy plastic effects, no clutter, no unnecessary decorative elements.
The final image should look like an authentic, professionally designed fictional music release created specifically for this artist, with a strong sense of identity and emotional storytelling.
Food magazine poster with oversized type and asymmetry
- Model: ChatGPT · image · editorial food poster
- Run on varg:
gpt_image_2_5(toolimage) — spec https://varg.ai/models/gpt_image_2_5.md - Post: @CaliraVal, 2026-05-08 — 23,828 views, 457 likes, 398 bookmarks
- Who: CaliraVal (11k followers), AI content creator
- Result: https://pbs.twimg.com/media/HHy2YADbIAASTyt.jpg (+2 more in the post)
- Inputs: one food photo
- Sections: Four paragraphs: layout goal with an avoid list, structure zones and grid, typography rules, cropping and framing
- Why it spread: 24k views and 398 bookmarks: a short art-direction brief (asymmetry, overflow, magazine rhythm) that pulls GPT Image out of the generic centered AI poster.
Designer-grade modern poster layout with strong visual hierarchy, asymmetric composition, intentional imbalance, editorial rhythm, and luxury fashion magazine aesthetics. Avoid generic AI poster style, centered layouts, boring placement, floating elements, cheap event-poster feel, or template-based composition.
Use a clear structure with hero zone, typography zone, breathing space, supporting detail area, and balanced graphic flow. Apply aggressive negative space, modular grid systems, layered visual rhythm, controlled chaos, cinematic cropping, and premium brand-level composition.
Typography must feel like a design element — oversized headlines, ultra-small metadata, cropped text, vertical or rotated typography, overlapping layers, edge-aligned placement, broken grid layouts, and strong size contrast. Create tension, movement, and visual flow instead of clean predictable alignment.
Use bold editorial cropping, unexpected framing, object overflow, and cinematic close-ups. Inspired by luxury fashion campaigns.
Ink silhouette typography poster with text in the neck
- Model: GPT Image 2 · image · typographic profile poster
- Run on varg:
gpt_image_2_5(toolimage, aspect 4:5) — spec https://varg.ai/models/gpt_image_2_5.md - Post: @HustleXR, 2026-08-28 — 12,201 views, 360 likes, 235 bookmarks
- Who: HustleXR (6.1k followers), AI creator posting GPT Image prompts
- Found on: freestylefly/awesome-gpt-image-2
- Result: https://pbs.twimg.com/media/HQyP7mcbAAATRtg.jpg (+3 more in the post)
- Inputs: none (text only)
- Sections: one paragraph template: subject, construction, text block, logo, background and format
- Why it spread: 12.2k views and 235 bookmarks: a four-slot template ([HUMAN], [FEATURE], [TEXT], [LOGO]) for stencil-ink side profiles with lettering fused into the body.
High-contrast black and white typographic portrait poster of [HUMAN], shown in side profile with [FEATURE]. Build the portrait with bold black silhouette blocks, sharp negative space, rough ink edges, fragmented stencil shapes, tiny editorial microtext, vertical typographic accents and expressive hand-drawn calligraphic marks. Integrate one large readable text block saying “[TEXT]” in 2 to 4 stacked lines, placed only inside the neck and body area, using oversized scribbled lettering that feels fused into the silhouette. Include a graphic design logo reading “[LOGO]” near the footer. Minimal off-white paper background, asymmetrical layout, cropped vertical composition, experimental editorial poster design, raw ink print texture, aspect ratio 4:5.
Watercolor-collage city poster with layout callouts
- Model: GPT Image 2 · image · watercolor collage travel poster
- Run on varg:
gpt_image_2_5(toolimage, aspect 3:4) — spec https://varg.ai/models/gpt_image_2_5.md - Post: @SimplyAnnisa, 2026-05-24 — 10,727 views, 277 likes, 170 bookmarks
- Who: SimplyAnnisa (16.7k followers), AI creator posting GPT Image prompts
- Found on: freestylefly/awesome-gpt-image-2
- Result: https://pbs.twimg.com/media/HJDPTY-XgAAA2F6.jpg
- Inputs: none (text only)
- Sections: format, scene, bulleted design elements by position, style
- Why it spread: 10.7k views and 170 bookmarks: a scene plus a corner-by-corner list of text blocks (title, coordinates, script, vertical lists) for a Kaohsiung poster.
Watercolor and collage travel poster for Kaohsiung, Taiwan, vertical layout, soft pastel and warm golden hour color palette.
Scene: Waterfront promenade at sunset with a modern light rail tram on the right, a person walking a bicycle on the left, and people strolling along the path. In the background: Kaohsiung skyline with the 85 Sky Tower and other skyscrapers, a harbor with a ferry boat, port cranes, and the Pier-2 Art Center building with murals. Seagulls flying in the sky, trees lining the walkway, calm water reflecting the light.
Design elements:
- Large text top left: 'KAO HSIUNG 2026', smaller text below 'Sea Breeze Slow Light Southern City'
- Top right: Vertical Chinese characters, 'KAOHSIUNG TAIWAN' in small caps below
- Bottom left: Coordinates '22.6273° N 120.3014° E', 'KAOHSIUNG TAIWAN', wave icon
- Bottom center: Handwritten script 'Kaohsiung'
- Bottom right: Vertical list 'Harbor Art Light Rail People Slow Life', vertical Chinese text
- Overlay effects: Semi-transparent geometric shapes, watercolor splashes, and texture layers in teal and beige
Style: Modern travel poster meets watercolor painting, airy and atmospheric, blend of illustration and photo-collage, warm and tranquil mood, high detail.
Glowing gold country silhouette poster on black
- Model: GPT Image 2 on OrcaRouter · image · country silhouette editorial poster
- Run on varg:
gpt_image_2_5(toolimage, aspect 9:16) — spec https://varg.ai/models/gpt_image_2_5.md - Post: @xRahultripathi, 2026-05-06 — 20,336 views, 198 likes, 115 bookmarks
- Who: xRahultripathi (4.5k followers), AI creator posting GPT Image prompts
- Found on: YouMind-OpenLab/awesome-gpt-image-2
- Result: https://pbs.twimg.com/media/HHqjWChbQAAO0aC.jpg (+3 more in the post)
- Inputs: none (text only)
- Sections: labelled blocks: format, STYLE, MAIN SUBJECT, INSIDE SILHOUETTE, LIGHTING, COLOR PALETTE, BACKGROUND, TYPOGRAPHY, FINISH
- Why it spread: 20.3k views: a labelled spec sheet (STYLE, MAIN SUBJECT, LIGHTING, PALETTE, TYPOGRAPHY) that fills any country outline with its skyline and landmarks in gold light.
4:5 vertical, ultra-high resolution (8K), 300 DPI, cinematic luxury editorial poster, gallery-grade realism, print-ready
STYLE:
Cinematic realism × glowing country silhouette collage × luxury magazine cover aesthetic × geopolitical visual storytelling
MAIN SUBJECT:
A glowing, semi-transparent silhouette of [COUNTRY NAME], centered in composition, radiating soft golden internal light, highly detailed geographical shape
INSIDE SILHOUETTE CONTENT:
Seamless cinematic integration of [LANDMARK], blended naturally inside the silhouette along with modern skylines, cultural identity elements, infrastructure, and futuristic city visuals
LIGHTING:
High-contrast cinematic lighting
Golden internal glow from within silhouette
Soft rim lighting on outer edges
Premium editorial highlight control
COLOR PALETTE:
Deep matte black background
Gold dominant glow inside silhouette
Natural realistic accent colors from landmarks
BACKGROUND:
Pure black cinematic void, minimal texture, luxury negative space
TYPOGRAPHY (OPTIONAL):
“THE RISING [COUNTRY NAME]”
Subtext: “Legacy. Strength. Future.”
FINISH:
Ultra-realistic, film grain, magazine print texture, clean composition, no cartoon look, no distortion, no artifacts
Prompt skills (0)
More for headlines & callouts
Effects (15)
- Kicker Over Rising Headline — HeyGen: A small mono kicker fades in, then the headline rises word by word from a soft blur.
- Rotating Benefit Word — Aceternity UI: A fixed phrase holds while one word in it cycles through benefits, uses or audiences.
- Masked Line Reveal — Daniel Petho: Each line slides up out of an invisible slot, so the type looks cut from the frame.
- Marker on the Key Phrase — HeyGen: A marker band draws behind the one phrase that matters while the rest of the line stays calm.
- Strike Old, Write New — HeyGen: A line strikes through the old way, then the new word takes its place.
- Clean Lower Third — HeyGen: Name and role slide in at the lower left on a quiet bar, hold 3–5 s, then wipe out.
- Hand-Drawn Callout Arrow — HeyGen: A wobbly arrow draws itself from a short label to the exact button or number on screen.
- Dim Everything But the Target — HeyGen: The frame darkens except a soft ellipse around the element being talked about.
- Impact Headline Slam — HeyGen: A heavy headline scales down hard into frame and lands with a 2–3 frame shake.
- Phrase Builds From the Center — Remocn · HeyGen: Words arrive one at a time and push the growing phrase so it always sits centered.
- Specular Sweep Title — HeyGen: A static title receives one clean light sweep through its glyphs, then rests.
- Decode-In Tech Title — HeyGen: Random glyphs churn and lock into the real word from left to right.
- Footage Inside the Letters — Magic UI: Huge block letters act as windows onto playing product footage or a demo clip.
- Tracking-In Feature Label — HeyGen: Wide-spaced letters draw together into a tight, crisp label as the blur clears.
- Title Design: lower thirds and title reveals by broadcast rules — varg: Lower thirds, speaker plates, show opens and glitch, CRT or chrome title reveals with hold times, safe areas and easing as presets.
Other kinds of skills
- Effects: An effect you can see, added to your video: animated titles, captions, transitions, shader backgrounds, logo animations, device mockups and viral camera tricks. Hand your agent the footage or the brief and get the effect.
- Styles: The look of the whole video: art styles, film looks, color grades and LUTs, caption fonts and style references. Pick a style and every shot your agent makes follows it.
- Guides: Know-how your agent reads before it works: how to write a script, storyboard, edit and caption, and how to use Remotion, HyperFrames, GSAP, Three.js and FFmpeg.
- Workflows: A brief in, a finished video out: UGC ads, product videos, music videos, talking heads, explainers, Shorts and short dramas, step by step, with the models and the edit chosen for you.
- Models & tools: Run one model or tool from your agent: Seedance, Veo, Kling, FLUX, GPT Image, Nano Banana, ElevenLabs and 1,200+ more, plus scrapers, trend search and video analysis.