Content & Media Creation · iart-ai/ad-video-skills
launch-video
This skill should be used when the user asks to "make a launch video", "create a product hype/sizzle film", "build a teaser/trailer for a drop", "do a reveal montage with an end-card CTA", "cut a hero video to the beat", or "export a launch film in 16:9/9:16/1:1". Covers the hook→tease→reveal→feature-montage→end-card arc and multi-aspect export.
L4 High-Privilege · Graded auth & full auditV1 Commit-PinnedExecute CommandNetwork AccessWrite FileDual-pass Audit
Control levels describe skill capabilities and suggested control measures; they do not represent quality or safety certifications, nor do they dictate usability. Usage decisions rest entirely with users and administrators. Different agent hosts apply varying constraints; do not assume hosts provide automatic protection. Third-party skill copyrights belong to their original authors. Use at your own risk.
1How It Works & Mechanism
Rule #1: Design sound first, then cut video to it; never edit visuals first and add music later.
skills/launch-video/SKILL.md
1. **Sound design leads picture.** Lock the track first, mark the beats and the drop, then cut visuals to those marks. Never score a finished cut — the reveal lands on the drop, not near it.
Elevates the soundtrack from post-production polish to the timeline's baseline.
The five-stage arc is fixed; for 15s/30s/60s, scale only the middle section—never stretch the hook.
skills/launch-video/SKILL.md
Scale the same proportions for 15s (tighter) or 60s (longer montage, never a longer hook).
The reference provides rescaled timing plans across three tiers.
Align drop frames using measured timecodes: on a paused GSAP timeline, use set for a hard cut precisely at the DROP moment.
skills/launch-video/references/launch-structure.md
const DROP = 9.0; // measured drop timecode (s)
In SKILL.md, the same constant is annotated as "measured from the track".
Montage discipline: one benefit per shot, 0.6–1.0s duration, identical enter/exit curves.
skills/launch-video/references/launch-structure.md
Montage discipline: keep ONE enter curve and ONE exit across all shots; vary content, never grammar.
Vary the content, keep the syntax consistent; fast cutting only reads as confident this way.
Multi-aspect ratios: build a 16:9 master first, then reframe shot-by-shot without letterboxing.
skills/launch-video/references/launch-structure.md
For 9:16, **reframe** (pan/scale per shot to keep the subject centered) — do not letterbox.
The reference includes safe zone metrics and export workflows for all three tiers.
Heavyweight workflows (Remotion rendering) must bake beats/drops into props; forbid GSAP or audio callback timing.
skills/launch-video/SKILL.md
If the launch film is built/rendered in Remotion, treat it as the heavy tier: register a `<Composition>` (+ zod `schema` + `defaultProps`), drive every value off `useCurrentFrame()` — no GSAP timelines / audio-event callbacks / `Date.now()` / `Math.random()` at render time (bake beat + drop timecodes into props).
Separates the tracks between lightweight HTML/GSAP and heavyweight Remotion.
End-card discipline: hold logo + tagline + CTA static for ≥2s, one single action, without competing against animations for attention.
skills/launch-video/SKILL.md
Hold the logo + tagline + CTA still and clean for ≥2s. No busy motion competing with the CTA; let the brand settle.
The closure principle: static holds build credibility.
Validation loop: sample three points (hook / reveal / end-card), and use shipped props rather than defaultProps.
skills/launch-video/SKILL.md
**Verify loop — render stills → inspect → encode.** Sample the arc, with the **shipped** props (real headline/tagline/CTA), not just `defaultProps`.
For multi-aspect exports, verify the master on each ratio first before batch encoding.
2Core Capabilities
01Five-stage structural template with 15/30/60s rescaling
0230s shot-by-shot storyboard script (four columns: visual, motion, audio, cut point)
03Drop-frame reveal formula (1.18→1.0 overshoot + white flash)
04Kinetic montage shot formula (keyword pop + one visual per shot)
05Multi-aspect safe zones and export workflow (reframe, not letterbox)
06Sound design workflow (track selection → map structure → edit mapping → beat syncing → final mixdown)
3External Dependencies
| Kind | Dependency / Endpoint |
|---|---|
| package | remotion（重档渲染 + zod 契约） |
| cli | npx remotion still / render |
| package | gsap（轻档 HTML 时间线） |
| cli | ffmpeg / ffprobe（仓库共享 contact-sheet.sh、probe-mp4.sh） |
| network | iart.ai（reference 尾部推广链接，具 utm 参数；仅读者点击才请求） |
4Recommended Governance & Evidence
Suggested Controls for Safe Operation
For Agent Runtime
Graded/per-use authorization, sandbox recommended
For Administrators
Approval workflow, rollback support, full audit
Evidence & Triggers (File Paths & Excerpts)
- scripts/: Executes local scripts or commands: No bundled scripts in the skill directory; SKILL.md instructs the agent to run `npx remotion still` / `npx remotion re
- 网络端点: Declares external network access: Does not query business endpoints at runtime; audio tracks are supplied by the user (the skill specifies no external music or TTS services).
- 本地工作目录: File I/O and artifact generation: Writes `out/f-hook.png`, `out/f-reveal.png`, `out/f-end.png`, `out/launch-*.m` into the user workspace.
Audit Notes: The trigger item is the classification rubric "Relies on third-party plugins, mirrors, or remote packages": the skill directory itself is pure Markdown (no scripts, credential reads, or env reads). However, the heavyweight delivery path requires `remotion` (+ zod contracts) and `@remotion/cli` fetched and executed from npm via npx, while the lightweight path requires the `gsap` package. Following the same evaluation criteria that marked `remotion-video` in the same repo/batch as Orange, this skill is marked consistently. **Explicitly non-applicable items**: No reading of process.env / API keys / tokens / cookies / keychains; no asset uploads to external generative services (audio tracks, product images, and logos remain entirely local); no TLS downgrade, `--no-verify`, sandbox disabling, or anti-automation bypass. Trigger conditions: Occurs only when actually rendering a video (or loading gsap to run a lightweight page); does not occur during storyboard drafting.
5Second-Pass Verification
- [discrepancy] SKILL.md 是否声明了 reference 未提供的实现 — Deliver 段要求 register `<Composition>` + zod schema + defaultProps，但唯一 reference（launch-structure.md）通篇是分镜、CSS/GSAP 与 ffmpeg 层面内容，**没有任何 Remotion 组件实现**（对比同仓 ad-creative 的 batch-ad-pipeline.md 有完整组件）——重档要由 agent 从零写。
- [ok] 「no GSAP / audio-callback timing」与正文 GSAP 代码是否自相矛盾 — 正文的 GSAP + `audio.addEventListener("play", ...)` 属轻档（HTML 交付物），Deliver 段的重档纪律被 "If the launch film is built/rendered in Remotion" 限定；属分档并存，但读者易误把事件驱动时间线用于视频导出（已记入 risks）。
- [ok] 示例帧号与 drop 时间码是否自洽 — `--frame=270` 注释为 reveal；30fps 下 270 帧＝第 9.0s，正好等于 reference 的 `const DROP = 9.0`。
- [ok] hook 时长纪律是否前后一致 — SKILL.md 的「hook 0–3s / Never lengthen the hook」与 reference 的「The hook never gets longer — only the montage scales.」同口径。
- [ok] 终卡时长与多比例验证顺序 — end card ≥2s 在 SKILL.md 输出清单、reference QA 清单、Deliver 前置条件三处一致；「先每个比例验 master 再编码」与同批其他 skill 同一口径。
- [ok] 凭证/外发面反查 — token 扫描零实质命中；音轨与素材均为本地文件；skill 未指名任何外部音乐/TTS/托管服务。
6Verdict & Scenarios
Best for: Brand and growth teams launching new products/features who need high-energy 15–60s teasers: requires drop-aligned reveals, fast-cut montages, full-placement delivery in 16:9/9:16/1:1, and acceptance of Remotion heavyweight projects.
Not for: Ad creative scenarios requiring batch data generation of N variants (use ad-creative-video); simple slide-style product explanations (use presentation/product-demo skills); projects with no audio track or an unwillingness to handle beat alignment.
Install Agent-Ready
① Verified Mirror 2026-09-28
Option A · Direct DownloadDownload launch-video.tar.gz
sha256:
Option B · Copy Guide for Agent056c60923ef4e60f…Agents automatically download and verify mirrors from this JSON guide.
② Upstream GitHub
GitHub Repo ↗
Pinned to commit 0de0f2a1c1.