Broll Remotion Build logo

Broll Remotion Build

Community
erduo1998-cell
broll-remotion-build

Build explicit Remotion canary representative samples as Lead, or own one explicitly assigned contiguous chapter through design, deterministic Composition authoring, viewing, and revision.

Overview

Publishererduo1998-cell
Repositoryerduo-broll-loop-engineering
Skill namebroll-remotion-build
Stars
228
Forks
44
Bundled files
5
LicenseMIT
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.

  • 5 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by erduo1998-cell on GitHub. Read the source before you install it.

Installation

Install the Broll Remotion Build 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/erduo1998-cell/erduo-broll-loop-engineering/tree/main/erduo-broll-loop-engineering/stages/broll-remotion-build
  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/erduo1998-cell/erduo-broll-loop-engineering.git /tmp/erduo-broll-loop-engineering
mkdir -p .claude/skills
cp -r /tmp/erduo-broll-loop-engineering/erduo-broll-loop-engineering/stages/broll-remotion-build .claude/skills/broll-remotion-build
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Broll Remotion Build 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 Broll Remotion Build 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 Broll Remotion Build 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.

Remotion Lead / Chapter Builder

Remotion is explicit opt-in or canary in v1.0.1, not the production default. Own only the assignment-selected role and source tree. Never change SRT timing, Recipe truth, another chapter, runtime selection, or Parent-generated media contracts. A failure never silently switches backend.

Complete creative context

Read the complete original SRT and original design directly; a summary cannot replace either. Use the assignment-selected role prompt, this assignment, its Recipes, chapter/seam context, shared asset/font index, Lead samples and capability index, exact project-local Remotion toolchain, and at most two selected references. Do not reopen Parent/other stage Skills, schemas, validators, generic craft references, backend guides, full catalogs, or command source.

truth is immutable. creativeProposal is revisable: change its metaphor, objects, composition, motion, material route, or key states when a better answer preserves the audience outcome and required readable result. Record one concise reason, not a new approval workflow.

Do not create src/inspection.tsx, inspection Compositions, DOM trace markers, motion windows, dense diagnostics, capture/trace/lint/screenshot/frame-scan/ hash/probe/decode/manifest/contract/receipt/proof tools, or passing proof artifacts. Parent owns deterministic rendering and media evidence.

Use only the assignment's exact local packages, CLI, profile, and prepared dependency identity. Never use global packages, download-style npx, a private per-unit install, HyperFrames execution, or a substituted profile. All visible motion derives from current frame/fps, deterministic math, or fixed seeds; no CSS time animation, timers, ambient state, network-at-render, wall clock, or unseeded randomness.

Use project-local font loading for every visible typeface named by the assignment; generic or host-system fallback is forbidden. Parent owns dependency preparation, typecheck, bundle reuse, direct shot rendering, and the fixed two-slot heavy-work queue. Parent ensures one dependency identity installs once per production root through remotion-toolchain.mjs prepare. Do not run a private per-unit dependency install. Parent invokes heavy commands through remotion-toolchain.mjs run-heavy.

Lead mode

Read the complete original inputs, compact motion map, exactly three representative Recipes, and shared assets/fonts. Build three final, directly renderable Compositions with clearly different silhouettes, compositions, and main actions:

  1. native graphic and type;
  2. real or generated material fusion;
  3. information-dense interface, process, or data organization.

Implement every signature motion named by the design in a real sample or runnable primitive. Provide reusable crop, mask, depth, background fusion, source binding, preparation, main action, overlap, settle, and readable-hold capabilities. The capability index stays under one page and explains which content relationship each capability serves. Shared source may contain tokens, visual/motion primitives, and base components, never a swap-the-copy whole-shot template.

Run only the exact standard command, open all three six-frame sheets and short previews, repair low-level defects, then record accepted or revised. A revised conclusion names the actual source change. These samples become final source for their shots; Chapter Builders do not rebuild them.

Chapter Builder mode

Own one contiguous chapter, normally 5–8 shots. Control its composition variation, material choices, pacing, and neighboring handoffs. Use Lead capabilities without copying Lead layouts. Never let three consecutive shots share the same layout skeleton, entry, and rhythm.

Choose native, provided, search, generate, or mixed separately for each shot. The shared freeze does not close shot-specific search/generation unless the packet records a real user, capability, authorization, or cost restriction. Add selected material to the provenance index and fuse it into crop, mask, path, palette, depth, annotation, or state.

Register one final, deterministic Composition/direct render entry per assigned shotId. Build the maximum visible result first, then staging, necessary anticipation, primary causal action, weaker follow-through, settle, and readable hold. Apply only the Recipe's 2–4 craftIntent principles; do not score or mechanically force all twelve. Keep stable holds valid and motion finite.

Run only the exact standard command. Open every shot's six-frame sheet and the chapter preview. Fix coverage, overlap, accumulated states, unsupported lines, empty containers, unreadable results, unfinished action, design-energy mismatch, and repeated chapter structure. Rerender only affected shots. Handoff requires one concise accepted or revised conclusion; a revision names what actually changed.

Complete only when every assigned shot has editable direct Composition source, Parent's technical command passes, the real sheets/chapter preview were viewed, truth is preserved, required results are readable, low-level defects are repaired, and material provenance closes. Technical success alone is not aesthetic approval.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Broll Remotion Build AI skill do?

Build explicit Remotion canary representative samples as Lead, or own one explicitly assigned contiguous chapter through design, deterministic Composition authoring, viewing, and revision.

Why use Broll Remotion Build on TypingMind?

Because you install it once and use it with any model. Broll Remotion Build 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 Broll Remotion Build in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/erduo1998-cell/erduo-broll-loop-engineering/tree/main/erduo-broll-loop-engineering/stages/broll-remotion-build. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Broll Remotion Build?

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 Broll Remotion Build?

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

Is the Broll Remotion Build AI skill free?

Yes. It is published on GitHub by erduo1998-cell under the MIT license. 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 👇