跳到正文
cccskills
免费在 GitHub 公开

image-to-video

FFmpeg-based video creation from image and audio.

中文简介正在准备中。当前显示原文说明。

查看安装方法

包含的文件(7)

  • SKILL.md3.2 KB
  • references/ffmpeg-filters.md4.9 KB
  • scripts/image_to_video.py17.2 KB
  • workspace/.gitignore202 B
  • workspace/completed/.gitkeep0 B
  • workspace/output/.gitkeep0 B
  • workspace/README.md1.4 KB

SKILL.md(原文)

安装前可以查看交给 AI 代理的指令内容。

Image to Video Skill

Combine a static image with an audio file to produce an MP4 via FFmpeg. Supports resolution presets, audio visualization overlays, and batch processing. For image generation, use image-gen.

Deep References

SignalLoadWhy
FFmpeg filter graphs for visualization modesreferences/ffmpeg-filters.mdScale/pad, showwaves, showspectrum, overlay filters

Phase 1: VALIDATE

  1. Check FFmpeg: ffmpeg -version. If missing, stop with install instructions.
  2. Verify both input files exist with absolute paths and non-zero size. Supported: PNG/JPG/JPEG/GIF/WEBP/BMP (image), MP3/WAV/M4A/OGG/FLAC (audio).
  3. Determine parameters from the user's request -- do not default to static when the user requested a visualization.
PresetDimensionsPlatform
1080p1920x1080YouTube HD (default)
720p1280x720Standard HD
square1080x1080Instagram, social
vertical1080x1920Stories, Reels, TikTok

Visualization modes (off unless requested): waveform, spectrum, cqt, bars.

Gate: FFmpeg installed, both files exist, parameters resolved.

Phase 2: PREPARE

Use the user's output path or derive from audio filename (/same/dir/filename.mp4). Verify directory is writable.

Phase 3: ENCODE

Defaults: libx264 preset medium, CRF 23, yuv420p, 192k AAC.

python3 $HOME/vexjoy-agent/skills/content/image-to-video/scripts/image_to_video.py \
  --image /path/to/image.png --audio /path/to/audio.mp3 \
  --output /path/to/output.mp4 --resolution 1080p --visualization static

Batch mode (matched pairs in workspace/input/):

python3 $HOME/vexjoy-agent/skills/content/image-to-video/scripts/image_to_video.py \
  --process-workspace --visualization waveform

Gate: Script exits 0.

Phase 4: VERIFY

FFmpeg can exit 0 but produce a corrupt file. Always probe:

ffprobe -v error -show_entries format=duration,size -show_entries stream=codec_name,width,height \
  -of default=noprint_wrappers=1 /path/to/output.mp4

Confirm video duration matches audio (within 1s). Report: file path, size, duration, resolution, visualization mode.

Error Handling

ErrorCauseSolution
FFmpeg not foundNot installedapt install ffmpeg or brew install ffmpeg
Image/audio not foundWrong or relative pathUse absolute paths; check with ls -la
FFmpeg filter errorsMinimal build lacks showwaves/showcqtInstall full FFmpeg; fall back to --visualization static
Cannot determine audio durationCorrupted audio fileTest with ffprobe; convert: ffmpeg -i input -acodec pcm_s16le output.wav

评价

还没有评价。欢迎分享你的使用感受。

评论

无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。

还没有评论。来写第一条吧。

提交后会立即公开。请勿填写邮箱、电话等个人信息。URL 最多 2 个。

0 / 1000 字

此表单受 reCAPTCHA 保护,适用 Google 的隐私权政策和服务条款。

同一仓库的技能

accounts

免费

Manage multiple Claude Code accounts: add, list, check, launch, and install shell aliases for 10+ isolated CLAUDE_CONFIG_DIR profiles.

中文简介正在准备中。当前显示原文说明。

notque/vexjoy-agent4422026年10月20日 更新

Improve architecture across modules by deepening interfaces.

中文简介正在准备中。当前显示原文说明。

notque/vexjoy-agent4422026年10月20日 更新

Assessment: read-only inspection, codebase overview, value analysis, health checks, ADR consultation, decision analysis, multi-perspective critique.

中文简介正在准备中。当前显示原文说明。

notque/vexjoy-agent4422026年10月20日 更新

Background memory consolidation — overnight review, merge, and injection payload for memory files.

中文简介正在准备中。当前显示原文说明。

notque/vexjoy-agent4422026年10月20日 更新

Jev-driven browser automation: Jev picks operations, programs execute, a text model writes field values only when Jev cannot pick one from the goal.

中文简介正在准备中。当前显示原文说明。

notque/vexjoy-agent4422026年10月20日 更新

Write, compose, integrate, and improve programs that call Jev, TypeSafe's System One judgment model.

中文简介正在准备中。当前显示原文说明。

notque/vexjoy-agent4422026年10月20日 更新

查看 notque 的全部技能

报告此技能的问题