cobalt-downloader

Download YouTube videos or audio streams via the Cobalt API.

25|5|Updated Dec 24, 2025
One-click install
npx skills add https://github.com/buda-ai/bunny-agent --skill cobalt-downloader
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cobalt-downloader
Source: https://github.com/buda-ai/bunny-agent/tree/main/templates/shortsdrone-agent/.claude/skills/cobalt-downloader
Command: npx skills add https://github.com/buda-ai/bunny-agent --skill cobalt-downloader

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Download videos from YouTube using the Cobalt API (self-hosted or public). No cookies required; more reliable than direct scrapers.

Core Features & Use Cases

  • Self-hosted or public deployment: download videos via Cobalt in a privacy-preserving way.
  • Video and audio options: download video streams or extract audio only.
  • Reliability: avoids YouTube anti-bot measures by leveraging Cobalt with a clear fallback path (yt-dlp if needed).

Quick Start

Point COBALT_API to your Cobalt instance and provide a YouTube URL to start the download.

Frequently Asked Questions about cobalt-downloader

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I download YouTube videos using the Cobalt API?

To download YouTube videos using the Cobalt API, provide your desired video URL and point the COBALT_API configuration to a running Cobalt instance. The API reliably fetches single or multiple video streams without requiring cookies.

Can I extract audio only from YouTube videos with Cobalt?

Yes, you can extract audio only from YouTube videos with Cobalt. The downloader includes an optional audio-only workflow with built-in error handling to fetch audio streams alongside standard video output options.

Do I need cookies to download YouTube videos via the Cobalt API?

No, you do not need cookies to download YouTube videos via the Cobalt API. Cobalt avoids YouTube anti-bot measures and direct scrapers, making it a more reliable privacy-preserving method for fetching streams.

What's the best way to self-host a downloader for YouTube videos?

The best way to self-host a downloader for YouTube videos is leveraging the Cobalt API on a self-hosted or public deployment. This avoids direct scrapers and cookies, providing reliable video and audio stream fetching.

Are there fallback options if YouTube video downloads stop working?

Yes, there is a clear fallback path using yt-dlp if YouTube video downloads stop working. This ensures reliable stream fetching even when default Cobalt API requests encounter anti-bot measures or errors.