magiccube/agentara/video-poster-design
Claude Code and agent skill by magiccube/agentara
- community
Checking that the SKILL.md file is still in the repository.
Create cinematic movie poster concepts and final poster images from a user's brief, existing video plan, storyboard, character design, or project context. Use this skill whenever the user asks to generate a film poster, movie poster, key art, campaign poster, teaser poster, title poster, or promotional visual for a video project. This skill proposes several creative directions first when appropriate, defaults to a 3:4 poster ratio when no ratio is specified, and ensures typography, title treatment, imagery, genre, setting, and story tone are coherent and visually inventive.
Install
npx vskill@latest install magiccube/agentara/video-poster-designSKILL.md
Video Poster Design
Use this skill to create movie-poster-style key art for a video, short film, vlog, TVC, storyboard, or campaign.
The final output is a poster image. When useful, also create a short poster spec or prompt file so the design can be regenerated consistently.
Default Behavior
If the user does not specify an aspect ratio, use 3:4.
If the user asks for a poster but has not chosen a creative direction, propose 3-5 poster concepts first and ask them to choose one. Keep the options concise and visually distinct.
If the user already gives a clear direction, title, ratio, and story context, proceed directly to image generation.
Output Location
Use a project-local poster directory:
Recommended file names:
Use lowercase, hyphenated, ASCII-safe names. Examples:
posters/youth-unlimited.pngposters/meaning-of-travel-teaser.pngposters/campus-energy-key-art.md
Do not overwrite existing files unless the user explicitly requests replacement. For variants, use suffixes such as -v2, -teaser, -main, -alt-a.
Inputs to Extract
Before designing, identify:
- Poster title and subtitle.
- Video/story context.
- Genre: drama, travel, romance, documentary, commercial, thriller, coming-of-age, sports, etc.
- Required aspect ratio. Default:
3:4. - Main character(s), product, place, or object.
- Character design images or approved reference assets.
- Storyboard or scene images that define visual continuity.
- Mood and emotional promise.
- Setting, time period, weather, and culture.
- Typography language: English, Chinese, bilingual, no text, etc.
- Required copy: tagline, credits, release date, brand line, campaign line.
- Output path if specified.
If the user does not specify text language, use the user's local language for any prompt/spec, and use poster typography that matches the title/campaign language.
Use Available Visual References
If the context contains any of these assets, use them as references for poster generation when relevant:
characters/*.pngcharacter design sheets.- Approved character
.mdspecs. storyboard/scene-XX.pngimages.- Existing poster drafts.
- Product or brand assets provided by the user.
When a character design image exists, feed it to the image generation model as a visual reference so the poster preserves character identity, hairstyle, wardrobe, body proportions, and material details.
When a storyboard image exists, use it to preserve scene language, color palette, setting, and story mood.
Creative Direction Options
When proposing options, make them concrete. Each option should include:
- Concept name.
- Main image composition.
- Typography direction.
- Color and lighting.
- Story emotion.
- Why it fits.
Example option format:
Do not make all options variations of the same composition. Vary the visual strategy:
- Character-led portrait.
- Environmental wide poster.
- Object-led symbolic poster.
- Motion/action poster.
- Minimal teaser.
- Ensemble composition.
- Graphic split or reflection concept.
- Product-as-hero commercial poster.
Typography Principles
Poster typography must match the story, setting, genre, and background image. It should not feel like generic overlay text.
Choose type direction based on context:
- Coming-of-age / campus: energetic but clean, slightly condensed sans serif, hand-painted or marker accent only if it feels natural.
- Travel documentary: editorial serif or restrained humanist sans, generous spacing, quiet placement.
- Romance: soft serif, handwritten accent, or understated title lockup depending on tone.
- Sports / youth TVC: bold condensed sans, dynamic italic or kinetic stacked layout.
- Thriller: severe condensed type, high contrast, restrained title placement.
- Sci-fi: technical geometry, but avoid generic neon unless the setting supports it.
- Period film: typography should reference the era without becoming costume-like.
- Minimal arthouse: sparse type, strong negative space, carefully placed title.
Integrate typography with the image:
- Let title placement interact with light, architecture, silhouette, product shape, or motion direction.
- Keep text legible.
- Avoid random fonts, fake credits, distorted letterforms, and unreadable generated text.
- If exact typography matters, consider generating poster art with clean space for text, then add text using deterministic design tools or image editing.
Poster Spec Template
When saving a .md spec, use:
Image Generation Prompt Structure
Use this structure for the final poster image:
Workflow
- Read relevant project context: plan, storyboard, character design, prior poster references, or user brief.
- Extract title, ratio, story, genre, main subject, visual references, and required text.
- If ratio is missing, set it to
3:4. - If the creative direction is not already clear, propose
3-5concepts and ask the user to choose. - After the user chooses, write a compact poster spec if helpful.
- Generate the poster image using the selected concept and all relevant references.
- Save the image to
posters/{poster-name}.png. - Inspect the result if possible: ratio, composition, typography, character consistency, artifacts.
- If text rendering is poor and exact text matters, recommend or perform a second pass with deterministic text overlay when appropriate.
- Final response should include the saved path and render the image when supported.
Quality Checklist
Before finalizing, verify:
- Aspect ratio matches the requested ratio or defaults to
3:4. - Poster concept fits the story, genre, background, and setting.
- Typography direction matches the narrative world.
- Title is legible enough for the intended use.
- There is no random or unintended text.
- Character identity is consistent when references are provided.
- Wardrobe and material details are consistent when character references are provided.
- Composition has a clear visual hierarchy.
- Poster does not look like a generic template.
- Output file is saved under
posters/unless the user requested otherwise.
Example
User asks:
If no direction is specified, propose:
- Run Toward the Light — full-body student running into morning track light, kinetic bold sans title.
- Blue Bottle Spark — product close-up with campus reflected in condensation, clean commercial typography.
- School Gate Launch — wide campus gate with students moving through sunlit trees, title integrated with shadows.
- After-Sweat Smile — character portrait with realistic sweat and blue-white palette, youth campaign title lockup.
Default ratio: 3:4.
Security
- Security scanPassed, 95 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckinguser-home/.claude/skills/video-poster-design/SKILL.md
Popularity
- Repository stars
- 462
- Repository forks
- 54
- 7-day trend score
- 31
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/magiccube/agentara/video-poster-design)