agentleFS
PiMPStudiosPiMPStudios/ACE-Step-1.5/.claude/skills/acestep-simplemv/SKILL.md

acestep-simplemv

Skill0 starsChanged 7 months agoMITView on GitHub

Render music videos from audio files and lyrics using Remotion. Accepts audio + LRC/JSON lyrics + title to produce MP4 videos with waveform visualization and synced lyrics display. Use when users mention MV generation, music video rendering, creating video from audio/lyrics, or visualizing songs.

  • Installs packages

3 other projects use a version of this, 1 of them edited.

Adopted by

Steady: 0 new projects took it in the last 4 weeks, none the 4 weeks before.thin data

Week of Sep 13: 0 new projectsWeek of Sep 20: 0 new projectsWeek of Sep 27: 0 new projectsWeek of Oct 4: 0 new projectsWeek of Oct 11: 0 new projectsWeek of Oct 18: 0 new projectsWeek of Oct 25: 0 new projectsWeek of Nov 1: 0 new projects
Sep 2026Oct 2026Nov 2026
New projects taking a version of this file each week, over the 8 weeks our latest run read: only 0 in all, so this is a hint, not a trend.

Most recently taken up by

The file

We split this skill into the rules it gives an agent

12 rules. About 0 are common, 5 are distinctive, and 7 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 7

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Before first use, check and install dependencies:

    line 18
  2. Distinctive:

    If ffprobe is not available, install ffmpeg (which includes ffprobe):

    line 31
  3. Very specific:

    If no browser is found, Remotion will attempt to download chrome-headless-shell from Google servers.

    line 82
  4. Very specific:

    When sending MV to chat platforms, use --max-size to auto-compress:

    line 122
  5. Distinctive:

    When running in containers (e.g.

    line 136
  6. Very specific:

    Detects: if CJK fonts are available (via fc-list)

    line 138
  7. Distinctive:

    If auto-install doesn't work, manually install fonts before rendering:

    line 142
  8. Very specific:

    IMPORTANT: Use the audio file's job ID as the output filename to avoid overwriting.

    line 156
  9. Distinctive:

    Default output uses the audio filename as base:

    line 158
  10. Very specific:

    Keep --title short and single-line (max ~50 chars, auto-truncated)

    line 167
  11. Very specific:

    Use --subtitle for additional info

    line 168
  12. Very specific:

    Do NOT put newlines in --title

    line 169

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

MV Render

Render music videos with waveform visualization and synced lyrics from audio + lyrics input.

Prerequisites

  • Remotion project at scripts/ directory within this skill
  • Node.js + npm dependencies installed
  • ffprobe available (for audio duration detection)

First-Time Setup

Before first use, check and install dependencies:

# 1. Check Node.js
node --version

# 2. Install npm dependencies
cd {project_root}/{.claude or .codex}/skills/acestep-simplemv/scripts && npm install

# 3. Check ffprobe
ffprobe -version

If ffprobe is not available, install ffmpeg (which includes ffprobe):

  • Windows: choco install ffmpeg or download from https://ffmpeg.org/download.html and add to PATH
  • macOS: brew install ffmpeg
  • Linux: sudo apt-get install ffmpeg (Debian/Ubuntu) or sudo dnf install ffmpeg (Fedora)

Quick Start

cd {project_root}/{.claude or .codex}/skills/acestep-simplemv/
./scripts/render-mv.sh --audio /path/to/song.mp3 --lyrics /path/to/song.lrc --title "Song Title"

Output: MP4 file at out/<audio_basename>.mp4 (or custom --output path).

Script Usage

./scripts/render-mv.sh --audio <file> --lyrics <lrc_file> --title "Title" [options]

Options:
  --audio        Audio file path (absolute paths supported)
  --lyrics       LRC format lyrics file (timestamped)
  --lyrics-json  JSON lyrics file [{start, end, text}] (alternative to --lyrics)
  --title        Video title (default: "Music Video")
  --subtitle     Subtitle text
  --credit       Bottom credit text
  --offset       Lyric timing offset in seconds (default: -0.5)
  --output       Output file path (default: out/<audio_basename>.mp4)
  --codec        h264|h265|vp8|vp9 (default: h264)
  --background   Background image file path (if omitted, uses animated gradient)
  --browser      Custom browser executable path (Chrome/Edge/Chromium)
  --max-size     Max output file size in MB (e.g. 24). Auto-compresses if exceeded.
                 Use for IM platforms (WhatsApp≤16MB, Discord≤25MB, Telegram≤50MB)

Environment variables:
  BROWSER_EXECUTABLE  Path to browser executable (overrides auto-detection)

Browser Detection

Remotion requires a Chromium-based browser for rendering. The script auto-detects browsers in this priority order:

  1. BROWSER_EXECUTABLE environment variable
  2. --browser CLI argument
  3. Remotion cache (chrome-headless-shell, downloaded by Remotion)
  4. System Chrome (auto-uses --chrome-mode=chrome-for-testing)
  5. System Edge (pre-installed on Windows 10/11, auto-uses --chrome-mode=chrome-for-testing)
  6. System Chromium (auto-uses --chrome-mode=chrome-for-testing)

Important: New versions of Chrome/Edge removed the old headless mode. When using regular Chrome/Edge/Chromium, the script automatically sets --chrome-mode=chrome-for-testing (which uses --headless=new). When using chrome-headless-shell, it uses the default headless-shell mode (which uses --headless=old). This is handled transparently.

If no browser is found, Remotion will attempt to download chrome-headless-shell from Google servers. This will fail if Google servers are inaccessible from your network.

Workarounds for restricted networks

Since Edge is pre-installed on Windows 10/11, it should be auto-detected without any manual configuration. The script automatically detects Chrome/Edge and uses the correct headless mode. If auto-detection fails:

# Option 1: Set environment variable
export BROWSER_EXECUTABLE="/path/to/msedge.exe"

# Option 2: Pass as CLI argument
./scripts/render-mv.sh --audio song.mp3 --lyrics song.lrc --title "Song" --browser "/path/to/msedge.exe"

# Option 3: Enable proxy and let Remotion download chrome-headless-shell

Examples

# Basic render
./scripts/render-mv.sh --audio /tmp/abc123_1.mp3 --lyrics /tmp/abc123.lrc --title "夜桜"

# Custom output path
./scripts/render-mv.sh --audio song.mp3 --lyrics song.lrc --title "My Song" --output /tmp/my_mv.mp4

# With subtitle and credit
./scripts/render-mv.sh --audio song.mp3 --lyrics song.lrc --title "Song" --subtitle "Artist Name" --credit "Generated by ACE-Step"

# With background image
./scripts/render-mv.sh --audio song.mp3 --lyrics song.lrc --title "Song" --background /path/to/cover.jpg

# Compress for Discord upload (max 25MB)
./scripts/render-mv.sh --audio song.mp3 --lyrics song.lrc --title "Song" --max-size 24

# Compress for WhatsApp (max 16MB)
./scripts/render-mv.sh --audio song.mp3 --lyrics song.lrc --title "Song" --max-size 15

IM Platform Upload Limits

When sending MV to chat platforms, use --max-size to auto-compress:

PlatformLimitRecommended --max-size
WhatsApp16MB10
Discord (free)25MB10
QQ10MB10
Telegram50MB20
Slack (free)1GB-

The compression uses ffmpeg two-pass encoding to achieve the best quality within the size constraint.

Container / Docker Font Support

When running in containers (e.g. OpenClaw), CJK fonts may not be pre-installed, causing lyrics to render as □ boxes. The script automatically:

  1. Detects if CJK fonts are available (via fc-list)
  2. Attempts to install fonts-noto-cjk (Debian/Ubuntu), font-noto-cjk (Alpine), or google-noto-sans-cjk-fonts (Fedora/RHEL)
  3. Falls back with a warning and manual install instructions if auto-install fails

If auto-install doesn't work, manually install fonts before rendering:

# Debian/Ubuntu
apt-get install -y fonts-noto-cjk

# Alpine
apk add font-noto-cjk

# Fedora/RHEL
dnf install -y google-noto-sans-cjk-fonts

File Naming

IMPORTANT: Use the audio file's job ID as the output filename to avoid overwriting. Do NOT use custom names like --output my_song.mp4. Let the default naming handle it (derives from audio filename).

Default output uses the audio filename as base:

  • Audio: acestep_output/{job_id}_1.mp3
  • Lyrics: acestep_output/{job_id}_1.lrc
  • Video: Pass --output acestep_output/{job_id}.mp4 (use the job ID from the audio file)

Example: if audio is chatcmpl-abc123_1.mp3, pass --output acestep_output/chatcmpl-abc123.mp4

Title Guidelines

  • Keep --title short and single-line (max ~50 chars, auto-truncated)
  • Use --subtitle for additional info
  • Do NOT put newlines in --title

Good: --title "Open Source" --subtitle "ACE-Step v1.5" Bad: --title "Open Source - ACE-Step v1.5\nCelebrating Music AI"

Notes

  • Audio files with absolute paths are auto-copied to public/ by render.mjs
  • Duration is auto-detected via ffprobe
  • Typical render time: ~1-2 minutes for a 90s song
  • Output resolution: 1920x1080, 30fps

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.