image-to-video-shots

80 / 100 · Low risk

Fine to try.

By xianyu110/ecommerce-video-skills (GitHub) · media · 131 stars · scanned at commit 96c49b3

主图转视频：把一张或几张商品主图/白底图/场景图，写成可直接喂给图生视频模型的镜头提示词（运镜、主体动作、光线、时长、负面约束），并保证商品不变形。Use when the user has product images and wants a product video, 主图视频, 图生视频, image-to-video, animate a product photo, or 让主图动起来 — with any image-to-video model (Seedance / 可灵 Kling / Veo / Runway / Hailuo / Wan …) or a local Ken Burns fallback.

Works with any agent that supports SKILL.md skills.

What this skill can do

Runs commands on your computer

The author does not mention this.

Found in 7 places.

skills/image-to-video-shots/SKILL.md, line 77python scripts/assemble.py storyboard.json -o out/main-video.mp4 --size 1080x1080 # 本目录已附带 assemble.py
skills/image-to-video-shots/scripts/assemble.py, line 43r = subprocess.run(cmd, capture_output=True, text=True)
skills/image-to-video-shots/scripts/assemble.py, line 141raw = subprocess.check_output(["ffmpeg", "-v", "error", "-i", path, "-f", "s16le", "-ac", "1", "-ar", str(SR), "-"])

Show 4 more

skills/image-to-video-shots/scripts/assemble.py, line 179r = subprocess.run(["ffmpeg", "-hide_banner", "-i", src, "-af",
skills/image-to-video-shots/scripts/make_subs.py, line 23names = subprocess.run(["fc-list", ":lang=zh", "family"], capture_output=True, text=True).stdout
skills/image-to-video-shots/scripts/tts_edge.py, line 23out = subprocess.check_output(["ffprobe", "-v", "error", "-show_entries", "format=duration",
skills/image-to-video-shots/scripts/tts_edge.py, line 51subprocess.run(["ffmpeg", "-y", "-v", "error", "-i", mp3, "-af", TRIM, "-ar", "48000", "-ac", "1", out_wav],

Downloads and installs other software

The author does not mention this.

Found in 2 places.

skills/image-to-video-shots/scripts/assemble.py, line 26Requires: ffmpeg (with libass), pip install pillow numpy edge-tts
skills/image-to-video-shots/scripts/tts_edge.py, line 13Needs: pip install edge-tts ; ffmpeg on PATH ; network access to the Edge TTS service.

Install

Paste this into your terminal. It downloads exactly the version we scanned, not whatever is newest.

if [ -e ~/.agents/skills/'image-to-video-shots' ] || [ -L ~/.agents/skills/'image-to-video-shots' ]; then echo 'A skill folder named image-to-video-shots already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/xianyu110/ecommerce-video-skills' "$d" && git -C "$d" checkout --quiet '96c49b349b7d975bda8f05ce6669c1d33500c269' -- 'skills/image-to-video-shots' && mkdir -p ~/.agents/skills && cp -R "$d"/'skills/image-to-video-shots' ~/.agents/skills/'image-to-video-shots' && rm -rf "$d"; fi

Codex reads ~/.agents/skills directly. For Claude Code, Cursor, Copilot, Gemini CLI, Antigravity or OpenCode, also link it. For Claude Code: ln -s ~/.agents/skills/'image-to-video-shots' ~/.claude/skills/'image-to-video-shots' (other agents: ~/.cursor/skills, ~/.copilot/skills, ~/.gemini/skills, ~/.gemini/config/skills, ~/.config/opencode/skills). Or install through the finder skill, which does both.

Badge for your README

The badge for xianyu110/ecommerce-video-skills shows the score of its lowest-scoring skill.

SkillTrust badge for xianyu110/ecommerce-video-skills

[![SkillTrust](https://zalavijaysinh.github.io/skilltrust/badges/xianyu110/ecommerce-video-skills.svg)](https://zalavijaysinh.github.io/skilltrust/repos/xianyu110/ecommerce-video-skills/)

Score history

Date	Version	Score
2026-11-07	`96c49b3`	80