Analyze video

Qwen Omni'yi kullanarak bir video veya görüntü dosyasını analiz edin

yazan kuscubasiece·8 Eylül 2026

Kaynak: davepoon/claude-code-subagents-collection · MIT

Kurulum

mkdir -p .claude/commands && curl -sL https://www.antrophiclove.com/api/kaynak/analyze-video/raw -o .claude/commands/analyze-video.md

Hedef dosya: .claude/commands/analyze-video.md

Önizleme Ham metin ⑂ Kendi sürümünü oluştur

Analyze a video or image file using the Qwen Vision Bridge.

Input

$1 — path to the video or image file
Remaining arguments ($ARGUMENTS minus $1) — optional analysis prompt. If not provided, use a general analysis prompt appropriate to the file type.

Steps

Verify the file exists at the given path. If not, search the current directory and workspace for a matching filename.
Check that DASHSCOPE_API_KEY is set:
```
echo $DASHSCOPE_API_KEY | head -c 5
```
If empty, tell the user to run /qwen-setup first.
Ensure dashscope is installed:
```
python3 -c "import dashscope" 2>&1
```
If missing, install it: pip install dashscope --break-system-packages

Run the bridge:

python3 "${CLAUDE_PLUGIN_ROOT}/skills/qwen-vision/scripts/qwen_bridge.py" "$1" "prompt here" --fps 2

Present the analysis to the user in a clean, readable format. If the user's prompt implies structured output (classification, comparison, scoring), format accordingly.
Ask if the user wants to:
- Follow up with more questions about the same video (use --save-context)
- Analyze another file
- Save the analysis to a file

For batch analysis

If the user provides a directory path or glob pattern instead of a single file, write a batch script following the pattern in ${CLAUDE_PLUGIN_ROOT}/skills/qwen-vision/references/batch-pattern.md and run it.