agentskills.codes
RE

remotion-solar-tts

by wustitesSource

Build or update narrated multilingual Remotion videos that combine Three.js solar-system visuals with Edge TTS voiceover. Use when Codex needs to scaffold, modify, validate, or render English, Chinese, Japanese, or Korean Remotion compositions with a Sun-to-planets narration timeline, generated MP3

Install

mkdir -p .claude/skills/remotion-solar-tts && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/13658" && unzip -o skill.zip -d .claude/skills/remotion-solar-tts && rm skill.zip

Installs to .claude/skills/remotion-solar-tts

Activation

This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.

Build or update narrated multilingual Remotion videos that combine Three.js solar-system visuals with Edge TTS voiceover. Use when Codex needs to scaffold, modify, validate, or render English, Chinese, Japanese, or Korean Remotion compositions with a Sun-to-planets narration timeline, generated MP3 narration, static Remotion audio assets, and smoke-test renders.
364 chars✓ has a “when” triggerlonger than Claude Code's old 250-char listing cap (fine on current versions)

About this skill

Remotion Solar TTS

Workflow

  1. Inspect the project shape with rg --files, package.json, src/Root.tsx, and src/SolarSystem.tsx.
  2. Keep the composition timeline and localized screen text in src/narration.ts; define fps, durationInFrames, language metadata, composition IDs, and ordered narration segments there.
  3. Keep visual sequencing deterministic. Drive camera motion, highlights, labels, and planet positions from useCurrentFrame() instead of timers or runtime randomness.
  4. Store voiceover source text in public/voiceover/narration.{lang}.txt.
  5. Generate narration with npm run voiceover, which runs scripts/generate_voiceover.py and writes ignored MP3 files such as public/voiceover/solar-system-en.mp3.
  6. Add each MP3 to the matching composition with Remotion <Audio src={staticFile(...)} />.
  7. Validate with npm run typecheck and npx remotion compositions src/index.ts.
  8. Render a still from at least one mid-sequence segment and a short video smoke test before considering the workflow complete.

Edge TTS

Prefer the project script over invoking edge-tts directly:

npm run voiceover

The script forces aiohttp through the Windows system DNS resolver because edge-tts can fail on Windows when aiodns cannot contact DNS servers.

If narration text changes, regenerate the MP3 files and compare each audio duration with the Remotion duration:

ffprobe -v error -show_entries format=duration -of default=noprint_wrappers=1:nokey=1 public/voiceover/solar-system-en.mp3
npx remotion compositions src/index.ts

Check each generated language MP3 and set durationInFrames to cover the longest duration plus a small tail.

Render Checks

Use these commands for fast confidence:

npm run typecheck
npx remotion compositions src/index.ts
npx remotion still src/index.ts SolarSystemEN out/solar-system-check.png --frame=780
npx remotion render src/index.ts SolarSystemEN out/solar-system-smoke.mp4 --frames=0-90

Check the smoke render has audio:

ffprobe -v error -show_streams -select_streams a out/solar-system-smoke.mp4

Use npm run render for all final language videos.

Search skills

Search the agent skills registry