storyboard-generator_skill

This skill polishes your story, splits it into storyboard frames, selects a compatible visual style, and generates consistent, ready-to-use storyboards.
  • Python

79

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill zephyrwang6/myskill --skill storyboard-generator

  • SKILL.md5.9 KB

Overview

This skill turns a user's story sketch into a polished storyboard sequence and batch-generates consistent storyboard illustrations. It refines the narrative, splits it into 6–10 storyboard panels with shot directions, and produces cohesive images in one of eight selectable visual styles. Generated images are saved to an Obsidian images folder and can be iterated on per-frame.

How this skill works

The skill first polishes the submitted story and extracts a concise story summary and emotional arc, then breaks the story into 6–10 panels with scene title, shot type, description, mood, and on-screen text. After confirming the panel script, it recommends 1–2 visual styles suited to the story, builds per-panel image prompts (style prefix, camera instruction, scene description, emotion, and an IMPORTANT consistency anchor), and invokes the Gemini image API to generate each frame. Images are saved to a dated Obsidian directory with consistent naming and resolution settings.

When to use it

  • You have a story outline and need a clear storyboard for production or pitching.
  • You want a quick set of cohesive concept images for a short film, animation, or comic.
  • You need style-consistent visuals for story pacing reviews or client presentations.
  • You want to iterate a single panel or re-generate failed frames without remaking the whole set.
  • You prefer automatic saving into an Obsidian image folder for easy project organization.

Best practices

  • Provide a clear prose description or bullet outline including main beats and key visual traits (character looks, locations, time of day).
  • Confirm or adjust the 6–10-panel split before generation to control pacing and content.
  • Select 1–2 recommended styles based on story tone; specify a preferred style to skip the recommendation step.
  • Include explicit character appearance details to ensure visual consistency across panels (hair, costume, body type).
  • Allow 2–3 second pauses between API calls to reduce rate-limit failures; review failure report and re-generate only failed frames.

Example use cases

  • Turning a short prose scene into a 7-panel storyboard for a director’s pitch.
  • Creating a 10-panel visual sequence in a Ghibli-watercolor style for a warm, coming-of-age short.
  • Generating a cyberpunk concept storyboard for a game pitch with neon-lit cityscapes.
  • Producing a black-and-white film-storyboard handoff for a commercial shoot.
  • Iteratively refining a single problem frame after client feedback while keeping other panels unchanged.

FAQ

Yes. If you specify a style (for example, Cyberpunk or Watercolor), the workflow skips recommendations and uses that style for all panels.

How are images named and where are they saved?

Images are saved to a dated Obsidian folder named MMDD-StoryName-storyboard with sequential filenames like 01-SceneShort.png, 02-SceneShort.png, etc., and a storyboard-script.md can also be saved.

What image settings are used by default?

Default outputs are 16:9 landscape, 2K resolution, and the Gemini image model; aspect ratio and resolution can be adjusted if needed.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational