Video enhancement
Collection of VergingAI Agent Skills
npx -y skills add verging-ai/agent-skills --skill video-enhancementAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
AI Video Enhancement - Upscale video resolution, denoise, sharpen. Supports local files and YouTube/Bilibili URLs. HD/4K upscaling with real-time progress.
SKILL.md
4.7 KB, ~1.3k tokens by cl100k_base, as published. Nobody here has run it
video-enhancement - AI Video Enhancement
Upscale and enhance video quality using AI via verging.ai.
Command Format
/video-enhancement --video <video file or URL> [options]
Options
| Option | Short | Description | Default |
|---|---|---|---|
| --video | -v | Video file path or URL (YouTube/Bilibili) | Required |
| --start | -s | Start time in seconds | 0 |
| --end | -e | End time in seconds | Video end (max 30s) |
| --hd | HD upscaling (5 credits/sec) | false (3 credits/sec) | |
| --api-key | -k | API Key | $VERGING_API_KEY |
| --output | -o | Save path for result | Current directory |
| --download | -d | Auto download result | false |
Authentication
Recommended: Authorization: ApiKey <your_key>
# ✅ Recommended (canonical form)
curl -H "Authorization: ApiKey vrg_sk_your_key_here" https://verging.ai/api/v1/auth/me
# ✅ Also works (Bearer with API key is supported)
curl -H "Authorization: Bearer vrg_sk_your_key_here" https://verging.ai/api/v1/auth/me
Get your API key: https://verging.ai → Login → Avatar → API Keys
API Reference (Exact Formats)
1. Check Credits
curl -H "Authorization: ApiKey $VERGING_API_KEY" \
https://verging.ai/api/v1/auth/me
Response: {"email":"...","name":"...","credits":100}
2. Get Upload URL (Form Data - NOT JSON)
# ⚠️ MUST use -F (multipart form-data)
curl -X POST https://verging.ai/api/v1/upload-video \
-H "Authorization: ApiKey $VERGING_API_KEY" \
-F "video_file_name=video.mp4" \
-F "job_type=video-enhance"
Response:
{
"code": 100000,
"result": {
"url": "https://...presigned-url...",
"public_url": "https://img.panpan8.com/video-enhance/..."
}
}
3. Upload Video to R2
curl -X PUT -T /tmp/verging-video-enhancement/trimmed.mp4 \
-H "Content-Type: video/mp4" \
"<presigned_url_from_step_2>"
4. Create Enhancement Job (Form Data)
# ⚠️ Uses -F (form-data), NOT JSON
curl -X POST https://verging.ai/api/v1/video_enhance/create-job \
-H "Authorization: ApiKey $VERGING_API_KEY" \
-F "target_video_url=<public_url_from_step_2>" \
-F "file_name=video.mp4" \
-F "user_video_duration=30" \
-F "job_type=video-enhance" \
-F "is_hd=false"
Response: {"code":10000,"result":{"job_id":"123"},"message":{"en":"Request Success","zh":"提交任务成功"}}
5. Poll Job Status
curl -H "Authorization: ApiKey $VERGING_API_KEY" \
"https://verging.ai/api/v1/jobs/list-jobs?job_ids=<job_id>"
Response: [{"id":123,"job_type":"video-enhance","status":"COMPLETED","progress":100,"result_url":"https://...","created_at":"...","updated_at":"..."}]
Status values: PENDING → PROCESSING → COMPLETED (or FAILED)
Execution Flow (Follow Exactly)
- Parse args → extract video path/URL, options
- Download remote video (if URL):
yt-dlp "URL" -o /tmp/verging-video-enhancement/input.mp4 - Get duration →
ffprobe -v error -show_entries format=duration -of default=noprint_wrappers=1:nokey=1 video.mp4 - Trim if needed (--start/--end or duration > 30s):
ffmpeg -i input.mp4 -ss <start> -to <end> -c:v libx264 -c:a aac /tmp/verging-video-enhancement/trimmed.mp4 - Check credits → GET /api/v1/auth/me (need duration × 3 normal or ×5 HD)
- Upload video → POST /upload-video (Form Data) → PUT to presigned URL
- Create job → POST /video_enhance/create-job (Form Data)
- Poll status → GET /jobs/list-jobs?job_ids=X every 5 seconds until COMPLETED
- Return/download result → show result_url, optionally curl download
Critical Notes
- Use
Authorization: ApiKey <key>(recommended) —Bearer <key>also works /upload-videouses Form Data (-F) — NOT JSON/video_enhance/create-jobuses Form Data (-F) — NOT JSON, no file upload needed- Max video duration: 30 seconds — trim longer videos first
- Temp directory:
/tmp/verging-video-enhancement/ - If yt-dlp unavailable, ask user to download video manually
Credit Consumption
| Mode | Cost |
|---|---|
| Normal | 3 credits/second |
| HD | 5 credits/second |
Privacy & Security
- Set key via env:
export VERGING_API_KEY="your_key" - Never expose API key in public repos
- Only process media you have rights to
- Temp files cleaned up after use
What ships with it: 1 file
1.4 KB alongside SKILL.md
- README.md1.4 KB
Gives 0 of the 12 instructions most video audio skills give in ~1.3k tokens
Counted across 622 of the 795 authors here whose files we hold, read 2026-08-07
- Read individual rule files for detailed explanationsin 21 of 622, across 10 files
- Render final videoin 13 of 622, across 6 files
- Use WAV PCM 16kHz mono audio formatin 12 of 622, across 3 files
- Use this skill when dealing with Remotion codein 11 of 622, across 4 files
- Save generated audio to a WAV filein 11 of 622, across 4 files
- Handle conversion errors gracefullyin 10 of 622, across 6 files
- Add captions to videos alwaysin 10 of 622, across 4 files
- Generate music from text descriptions using MusicGenin 9 of 622, across 2 files
- Do not skip pipeline layersin 9 of 622, across 3 files
- Do not make one tool do everythingin 9 of 622, across 3 files
- Use Azure Document Intelligence for complex PDFsin 9 of 622, across 4 files
- Never ask the user to paste their full API keyin 9 of 622, across 3 files
Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.