ba0gu0/cobalt-download
skills/cobalt-download/SKILL.md
Download videos and audio from 20+ platforms (YouTube, TikTok, Instagram, Twitter/X, Reddit, Bilibili, Pinterest, SoundCloud, Vimeo, etc.) using the cobalt API. Auto-detects python/curl+jq/powershell backend. Use when: user wants to download a video, extract audio from a video, save a clip, or rip media from a social platform URL. Triggers: "download video", "下载视频", "save video", "rip audio", "extract audio", "download from youtube", "download from tiktok", "download from twitter", "下载", "保存视频", "提取音频".
tools
Updated May 13, 2026
$ install --global
skillsauthnpx skillsauth add ba0gu0/skillbox cobalt-downloadInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
Security Scan Results
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
3
Scanners Passed
9
Scanners in report
Clean
TrivyContainer and dependency vulnerability scanner
95%
Clean
SemgrepStatic code analysis for vulnerabilities
95%
Clean
mcp-scan (Snyk)Model Context Protocol security validation
95%
Skipped
Snyk (dep)Open source security scanning
50%
Skipped
Socket.devSupply chain security analysis
50%
Skipped
VirusTotalMulti-engine malware detection
50%
Skipped
CrowdStrikeAdvanced threat intelligence
50%
Skipped
OSV-ScannerOpen Source Vulnerability database check
50%
Skipped
OWASP Dep-Check
50%
Last scanned: May 13, 2026, 3:37 AM172.3s4 files scanned
SKILL.md
- name:
- cobalt-download
- description:
- >
- Use when:
- user wants to download a video, extract audio from a video,
- Triggers:
- download video", "下载视频", "save video", "rip audio",
Cobalt Download Skill
Download videos and audio from 20+ platforms via cobalt API. Zero-config — each run fetches the latest instance list from cobalt.directory, tries up to 10 instances, skips auth-required ones automatically.
Scripts
- Python:
scripts/cobalt-dl.py - Shell:
scripts/cobalt-dl.sh(curl + jq) - PowerShell:
scripts/cobalt-dl.ps1(Windows)
Quick Reference
# Download video (default 1080p)
python3 scripts/cobalt-dl.py "<url>"
# 4K video
python3 scripts/cobalt-dl.py "<url>" -q 2160
# Audio only (MP3 320kbps)
python3 scripts/cobalt-dl.py "<url>" --audio
# Audio as opus
python3 scripts/cobalt-dl.py "<url>" --audio -f opus
# Output directory
python3 scripts/cobalt-dl.py "<url>" -o ~/Downloads
# With subtitles
python3 scripts/cobalt-dl.py "<url>" --subs en,zh
Shell version (identical interface):
scripts/cobalt-dl.sh "<url>"
scripts/cobalt-dl.sh "<url>" --audio
PowerShell version (Windows):
pwsh scripts/cobalt-dl.ps1 "<url>"
pwsh scripts/cobalt-dl.ps1 "<url>" --audio
pwsh scripts/cobalt-dl.ps1 "<url>" -q 2160 -o C:\Downloads
Options
| Flag | Description | Default |
|------|-------------|---------|
| -q, --quality | Video quality (144-4320) | 1080 |
| -o, --output | Output directory | current dir |
| --audio | Audio only mode | off |
| -f, --format | Audio format (mp3/opus/ogg/wav) | mp3 |
| -b, --bitrate | Audio bitrate (64-320) | 320 |
| --mute | Video without audio | off |
| --subs | Subtitle languages (ISO 639-1) | none |
| --codec | YouTube codec (h264/av1/vp9) | h264 |
| --tiktok-full-audio | TikTok original sound | off |
Supported Platforms (22)
bilibili, bluesky, dailymotion, facebook, instagram, loom, ok, pinterest, newgrounds, reddit, rutube, snapchat, soundcloud, streamable, tiktok, tumblr, twitch clips, twitter/X, vimeo, vk, xiaohongshu, youtube
How It Works
- Fetch instance list from
cobalt.directory/api/working?type=api - If API unreachable, fall back to hardcoded instance list
- POST the URL to the first instance
- If auth required → skip to next instance (up to 10 tries)
- On success: download via requests (Python) / curl (Shell) / Invoke-WebRequest (PowerShell)
Response Types
tunnel/redirect→ single file, download viaurl+filenamepicker→ multiple items (e.g., Instagram carousel), iteratepicker[]error→ checkerror.codefor reason
Related Skills
ba0gu0/jina-web
development
VerifiedTrustedCommunity
Read web pages and search the internet using Jina AI Reader & Search APIs. Two modes: (1) READ - fetch any URL and convert to clean markdown/text, (2) SEARCH - web search with results returned as markdown. Supports proxy, CSS selectors, screenshots, caching control, and more. Auto-detects python/curl+jq/powershell backend. Use when: user wants to read a webpage, fetch URL content, scrape a site, search the web, or get clean text from a URL. Triggers: "read this page", "fetch URL", "scrape website", "search the web", "jina search", "jina read", "读取网页", "抓取页面", "搜索网页", "网页内容", "获取链接内容".
SKILL.mdUpdated May 13, 2026
ba0gu0/agent-ready-framework
development
VerifiedTrustedCommunity
项目接手系统生成器。把新项目、代码无文档项目或老项目升级成任何 AI Coding Agent 都能不靠聊天历史直接接手执行的工作系统。核心建立 5 个能力：项目说明、当前状态、下一步任务、验收标准、红线规则。
SKILL.mdUpdated May 13, 2026
openclaw/taskflow
tools
VerifiedTrustedCommunity
Use when work should span one or more detached tasks but still behave like one job with a single owner context. TaskFlow is the durable flow substrate under authoring layers like Lobster, ACPX, plugins, or plain code. Keep conditional logic in the caller; use TaskFlow for flow identity, child-task linkage, waiting state, revision-checked mutations, and user-facing emergence.
357,764SKILL.mdUpdated Apr 10, 2026
openclaw/extensions/lobster
tools
VerifiedTrustedCommunity
# Lobster Lobster executes multi-step workflows with approval checkpoints. Use it when: - User wants a repeatable automation (triage, monitor, sync) - Actions need human approval before executing (send, post, delete) - Multiple tool calls should run as one deterministic operation ## When to use Lobster | User intent | Use Lobster? | | ------------------------------------------------------ | --------------------------
357,764SKILL.mdUpdated Apr 10, 2026