Open Source
Open Source

video-shotcraft: Turn Claude Code Into a Cinematic Product Video Studio (3.5K Stars)

video-shotcraft (Vincentwei1021/video-shotcraft, 3.5K GitHub stars, TypeScript, Apache-2.0) is an agent skill that turns Claude Code or Codex into a motion-design studio - point it at your product and it storyboards, animates, and sound-designs a cinematic promo with Remotion. Ships 104 shot recipe cards, 161 motion previews, a validated 36.2s Ink Press template, 2.5D camera moves, beat-synced cuts, and film-grade SFX. Output is reproducible TSX, not black-box generation. Includes a China AtomGit mirror and 3 headless-rendering CI flags.

Published August 4, 20268 min read
<!-- video-shotcraft-resource | resource | video-shotcraft: Turn Claude Code Into a Cinematic Product Video Studio -->

Your product is done and you need a promo video that doesn't look amateur. The traditional path is hiring a designer to hand-craft it in AE/Pr, or stamping one out from a template site - the former is expensive and slow, the latter looks like everyone else's. video-shotcraft offers a third path: it's an agent skill that, once installed in Claude Code or Codex, lets you point at your product and have it storyboard, animate, and sound-design a cinematic promo rendered with Remotion.

What it is

video-shotcraft (github.com/Vincentwei1021/video-shotcraft) is an open-source project by developer Vincentwei1021 (Yihao) with 3,492 GitHub stars, 295 forks, primary language TypeScript, Apache-2.0 license, created 2026-07-19, last pushed today (August 4). One-line pitch: turn Claude Code or Codex into a motion-design studio - point it at your product and it storyboards, animates, and sound-designs a cinematic promo, marketing, launch, or demo video with Remotion.

It belongs to the same "agent skill" wave as book-to-skill - following the open Agent Skills standard, installed into an agent's skills directory, invoked on demand. But the direction differs: book-to-skill solves "reading," video-shotcraft solves "making video." The technical base is Remotion - an open-source framework for writing videos in React/TypeScript - so the output is programmable, reproducible TSX, not a black-box generation.

Four core capabilities: real page captures, 2.5D camera moves, beat-synced cuts, and film-grade SFX.

What's included

The repo ships not just a template but a full video-production asset library:

ContentDescription
104 shot recipe cardsPurpose, energy, suggested duration, parameters, implementation notes, known pitfalls
161 motion previewsCovering 161 styles; searchable and filterable in the online Gallery
Remotion implementationsTuned TSX demos with real easing and timing parameters
Complete video templateA validated 36.2-second, 1920x1080, 30fps, 10-shot product promo
Components and assets2.5D page camera, captions, flash cuts, digit rolls, SFX, capture scripts
Production methodologyCapture, visual direction, storyboarding, sound design, beat sync, final QA

The shot recipe cards are the core asset: each card is a reusable shot pattern (e.g. spotlight-hero-card, deck-deal-flyin, row-embed) with parameters and implementation notes that an agent can directly call and combine. The 161 motion previews are searchable, filterable, and variant-switchable in the online Gallery - pick a card, copy its name, hand it to the agent.

Built-in template: Ink Press

Don't want to assemble shots from scratch? The repo ships a validated complete template, Ink Press: 36.2 seconds, 1920x1080, 30fps, 10 shots, paper-ink-amber style, 2.5D real-page camera moves, with title cards, transitions, and a full cinematic SFX pass. One-line usage:

text
Use video-shotcraft to make a promo for my product with the Ink Press template.

The agent swaps in your product's screenshots, copy, and branding to reproduce the same quality. It's the fastest, most reliable path from zero to a finished film. More templates are on the way.

How to install and use

The most direct way: hand the repo link to your agent and say:

text
Install this skill for me: https://github.com/Vincentwei1021/video-shotcraft

The agent clones the repo and links it into your skills directory. Or use the skills CLI or do it manually:

bash
npx skills add Vincentwei1021/video-shotcraft
# or manually
git clone https://github.com/Vincentwei1021/video-shotcraft.git
ln -s "$(pwd)/video-shotcraft" ~/.claude/skills/video-shotcraft   # Claude Code
ln -s "$(pwd)/video-shotcraft" ~/.codex/skills/video-shotcraft    # Codex

Then make requests like:

text
Use video-shotcraft to create a promo for my desktop product.
Use the deck-deal-flyin and row-embed shot cards to present this feature.
Design a product close-up inspired by spotlight-hero-card.

If no shot card is specified, the skill introduces the built-in template first and asks whether to use it; you can also pick shots in the Gallery before starting. For developers in China: the author maintains an AtomGit mirror (atomgit.com/VincentWei/video-shotcraft) if GitHub is hard to reach.

Three headless rendering pitfalls

Rendering on a headless Linux box (CI) hits three known walls:

  1. Concurrency cap: remotion still/render fails with "Maximum for --concurrency is 2" on low-core machines. Fix: pass --concurrency=1.
  2. Old Headless removal: recent Chrome/Chromium dropped old headless mode; pointing Remotion at system chromium fails to launch. Fix: use a chrome-headless-shell binary instead of full Chrome.
  3. Blocked CDN: if remotion.media is unreachable, the automatic headless-shell download is rejected. Fix: --browser-executable=<path-to-local chrome-headless-shell>.

With these three flags set, frame renders from the bundled template work.

Takeaway

video-shotcraft hits the high-frequency, non-trivial need for "product promo videos." Its approach isn't to duke it out with generative video models like Sora, but to distill professional video production - shot patterns, motion parameters, audio beats - into a reusable agent skill asset library that an agent assembles with Remotion, a programmable framework. The output is TSX source code: readable, editable, reproducible - its biggest difference from black-box generation.

It suits two crowds: indie devs and small teams who've finished a product and need a presentable promo but can't afford video production; and people doing scaled AI-agent video production who need a reusable shot system. The bar is using a host that supports Agent Skills (Claude Code or Codex) and being able to install Remotion locally (Node and Chrome environment). The risk is that Remotion rendering is resource-hungry and headless environments are tricky - get the three flags ready for CI.

In the trend of agent skills extending from "reading docs" to "making creations," video-shotcraft is a template for distilling professional video-production methodology into an open-source skill. 3,492 stars, Trendshift-listed, pushed today - developers are clearly using it.


References

This article is AI-assisted and human-edited. Last updated: 2026-08-04

Related

Open Source

book-to-skill: Turn Any Technical Book Into an AI Agent Skill (16.2K Stars)

book-to-skill (virgiliojr94/book-to-skill, 16.2K GitHub stars, Python, MIT) distills technical books/PDFs/EPUBs/doc folders into structured agent skills following the open Agent Skills standard - install once, works across GitHub Copilot CLI, Amp, and Claude Code. Generates SKILL.md + per-chapter files + glossary + patterns + cheatsheet, with chapters loaded on-demand so they don't count against your token budget. Ships a benchmark tool measuring 24-51x fewer tokens than dumping the full book into context (tested on 3 real books). Beyond books: internal docs, brand systems, research clusters, specs. Includes a copyright-compliance note.

Aug 4, 20267 min read
Open Source

VoiceStudio: the local-first open-source voice studio

The GitHub repo debpalash/VoiceStudio gained +5104 stars in a single week (week of 2026-09-07) to about 24.6k total, topping that week's momentum charts as a local-first voice project (AGPL-3.0, Python, active on 2026-09-11). Its positioning fits one line: an open-source, fully local ElevenLabs alternative - voice cloning, voice design, video dubbing, dictation, transcription, audiobook creation, covering about 646 languages, with the local workflow needing no account, API key, subscription, or usage meter. The underrated design is that it is not one voice model but an engine-orchestration layer integrating 16 TTS and 11 ASR engines, hot-swappable; it runs across macOS/Windows/Linux/Docker and ships an OpenAI-compatible local speech API plus an MCP server. This piece maps the capability surface, the local-first privacy/cost divide, and the division of labor with the same-week cloud real-time GPT-Live-1 (VoiceStudio leans to batch dubbing/transcription, not real-time conversation), then names five real constraints: AGPL-3.0 commercial caveats, beta stability, the ongoing Electron rewrite, uneven engine quality, and not every engine being local or free.

Sep 13, 202610 min read
Open Source

LLaDA-Image: Ant Full-Open 6B Unified Image Generation Model

Ant Group's InclusionAI open-sourced LLaDA-Image, a 6B unified image generation and editing model (208 stars / Python / created 2026-08-31, snapshot 2026-09-09). One checkpoint does both text-to-image and instruction-guided editing; both backbone and DiT are diffusion models trained in a unified framework, with image-only pre-training establishing the visual prior; the Turbo variant uses Twin-DMD distillation to cut 50 steps down to 4. It scores 53.53 (English) and 53.38 (Chinese) on Qwen-Image-Bench, a double SOTA. HuggingFace and ModelScope host Base and Turbo weights, each with an FP8 variant, and community ComfyUI support landed on 2026-09-07. Biggest caveat: the repo's license field is null with no LICENSE file - confirm terms with InclusionAI before commercial use rather than assuming Apache-2.0 or MIT.

Sep 9, 202610 min read