<< All versions

Skill v2.0.0

currentAutomated scan100/100
rsclaw-ai/rsclaw/web-video-download

+3 new

──Details
PublishedSeptember 10, 2026 at 08:56 AM
Content Hashsha256:e933a8f040321d88...
Git SHAc1c53ae14e27
Bump Typepatch
Compare with v1.0.0
──Files
Files (1 file, 1.7 KB)
SKILL.md1.7 KBactive
SKILL.md · 60 lines · 1.7 KB

name: web-video-download description: Universal video download — capture video URL from any site and download with cookies version: 2.0.0 icon: "🎬" author: "@rsclaw"


Video Download

3 steps only. Works with Douyin, Bilibili, Kuaishou, Xiaohongshu, YouTube, TikTok, etc.

Usage notice / 使用声明
Intended for content you own or are authorised to access — your own
uploads, licensed material, or content the rights holder permits you to
save. You are responsible for complying with each platform's terms of
service and with applicable copyright law. Do not redistribute
third-party content without permission.
本技能仅用于下载您自有或已获授权访问的内容。请遵守各平台服务条款及著作权
法律法规;未经许可请勿转载或传播他人作品。

Step 1: Capture video URLs

json
{"tool": "web_browser", "action": "capture_video", "url": "<video_page_url>"}

This opens the page, waits for video to load, and returns all detected video URLs.

Step 2: Download

Pick the best URL from the result (prefer mp4/playaddr, avoid thumbnail/poster), then:

json
{"tool": "web_download", "url": "<best_video_url>", "path": "video.mp4", "use_browser_cookies": true}

path is just a filename — no ~/ or absolute paths.

Step 3: Send to user

json
{"tool": "send_file", "path": "<downloaded_file_path>"}

If login required

Use web_browser to screenshot the QR code and send to user:

json
{"tool": "web_browser", "action": "screenshot"}

Wait for user to scan, then retry step 1.

Rules

  • Do NOT use exec/curl/wget/yt-dlp
  • Always use use_browser_cookies: true in web_download
  • If capture_video returns empty, the video may need login or is DRM-protected
← v1.0.0All versions