Short Video Studio logo

Short Video Studio

CommunityPopular
Devin-AXIS
short-video-studio

Plan and create short videos using the installed short-video-studio workspace app. Manage its creative canvas, scripts, storyboards, characters and local assets, generate images or video through connected Work channels, and hand off editable tracks to Work Video Studio.

Overview

PublisherDevin-AXIS
RepositoryiPolloWork
Skill nameshort-video-studio
Stars
6.9K
Forks
1.1K
Bundled files
Instructions only
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by Devin-AXIS on GitHub. Read the source before you install it.

Installation

Install the Short Video Studio AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/Devin-AXIS/iPolloWork/tree/main/examples/plugin-packages/short-video-studio/skills/short-video-studio
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/Devin-AXIS/iPolloWork.git /tmp/iPolloWork
mkdir -p .claude/skills
cp -r /tmp/iPolloWork/examples/plugin-packages/short-video-studio/skills/short-video-studio .claude/skills/short-video-studio
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Short Video Studio in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Short Video Studio on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Short Video Studio is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Short-video workspace

This skill belongs to the independently installed short-video-studio plugin. Do not install a global copy or modify the iPolloWork source repository.

Conversation and workspace

The user talks in the existing left conversation. The right workspace is a creative canvas, not a second chat. Open the plugin's studio workspace app through Work's existing plugin entry. If already open, discover its tools using ipollowork_workspace_app_list_tools and call them with ipollowork_workspace_app_call_tool. Do not guess a browser URL or start another editor server.

  1. Call project_read. The project has nodes, edges, script, shots, roles, assets, tracks, jobs, exports, and a current revision.
  2. Propose a coherent story and update the same document with project_update. Preserve unrelated work and the current revision. If a revision conflict is returned, re-read and reconcile; do not blindly overwrite.
  3. Nodes have id, kind (text, image, video, audio, script, storyboard), title, prompt, x, y, optional assetId and settings. IDs start with a letter and contain only ASCII letters, numbers, _ or -. Edges are {id, from, to}; they organize the canvas, but are not themselves model references.
  4. Shots are {id, title, prompt, narration, duration, assetId?} in playback order. Duration is seconds. Roles are {id, name, description, assetId?}. Character consistency is a prompt/reference workflow, not a guarantee.

Real generation

Call capabilities before choosing a model. Use only models available/configured in Work and only the returned parameter choices. The plugin neither connects to LibTV nor stores another API key.

Image node settings: model, size, quality; optional referenceId for an image edit, maskDataUrl for an explicitly supplied transparent PNG mask. Video node settings: model, operation, resolution, duration (string), ratio; referenceId for first frame, lastFrameId for last frame, or referenceIds for multimodal references. First-frame, edit and extend modes use adaptive ratio; edit uses duration -1. style and camera are prompt intent, not native deterministic controls.

Audio nodes synthesize speech through Work's configured CosyVoice channel, using their prompt as narration text and optional settings.voice. The service measures the saved MP3 duration; place it on a separate audio track without truncating speech. This is not a music-generation model.

Call generate only when the user asks to generate. It can charge the configured provider. Pass a fresh UUID requestId for a new intended generation; reuse the same request ID after an uncertain transport result. Query generation_status; never automatically resubmit an uncertain task. Unsupported models, music generation, proprietary LibTV services and unavailable authorizations must be explained, not simulated.

Import user-selected local media via import_asset with a workspace-relative path. Assets become project-owned copies. Supply metadata.durationSeconds for media whose duration isn't available from Work inspection. Don't invent a duration or use remote expiring URLs in final tracks. Generated/imported assets, jobs, and exports are maintained by the service; project_update cannot manufacture them.

Editable multitrack handoff

Assign image/video assets to shots, then arrange_shots. This replaces rough-cut tracks, so ask before discarding existing manual track edits. Original sound becomes a separate audio track, not mixed into the video. Add music/voice/SFX as separate audio tracks; captions remain editable text.

Tracks contain {id, label, kind, clips}. Clip start/duration use integer startFrame/durationFrames at the project's rational fps. Media clips use source: assets/<registered filename>, sourceDurationSeconds, sourceStartSeconds, playbackRate, volume. Same-track overlap and source overruns are rejected. Never hand off a flattened MP4 as the editable project.

Call handoff. Return the resulting absolutePath or workspace-relative path as a clickable local file link in the left conversation and tell the user to open it in Work's Video Studio. The current host has no generic plugin API that switches directly into the native editor; do not claim it switched automatically. Once handed off, the native HTML and assets are the editing source of truth. Never overwrite them from an older canvas document. Changed handoffs create a new engineering file; repeated unchanged handoffs reuse the existing one.

Projects live under the chosen workspace's short-video/<project-id>/; native handoffs under video/<id>/. These user-created files survive plugin uninstall. The package, skill and temporary upload data are removed by the host's plugin lifecycle.

Frequently asked questions

What does the Short Video Studio AI skill do?

Plan and create short videos using the installed short-video-studio workspace app. Manage its creative canvas, scripts, storyboards, characters and local assets, generate images or video through connected Work channels, and hand off editable tracks to Work Video Studio.

Why use Short Video Studio on TypingMind?

Because you install it once and use it with any model. Short Video Studio is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Short Video Studio in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/Devin-AXIS/iPolloWork/tree/main/examples/plugin-packages/short-video-studio/skills/short-video-studio. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Short Video Studio?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Short Video Studio?

As many as you like. As long as a model supports skills, you can use Short Video Studio with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Short Video Studio AI skill free?

It is published on GitHub by Devin-AXIS. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇