yt-dlp-mcp
The MCP server gives agents video search, metadata, subtitles and authorized media downloads through local yt-dlp.
- 300 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- The server contacts external sites and writes downloads; configured cookies grant access through the user’s session.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add mcp/yt-dlp-mcpDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The server contacts external sites and writes downloads; configured cookies grant access through the user’s session.
This is third-party code. Review the repository files before installing.
What it does
yt-dlp-mcp connects agents to yt-dlp. Agents search videos, read metadata and obtain subtitles or clean transcripts. Authorized downloads can save video, audio or a selected segment locally. Results use JSON or Markdown, including threaded comments. Metadata does not require downloading the video. Users choose language, resolution and output directory for the task.
Who it is for. Video editors needing subtitles, metadata and local copies of authorized source material in agent workflows.
Good fit when
- A review needs title, date and description without downloading.
- Agent analysis needs an available transcript.
- Your own video needs a segment for further editing.
- Discussion research needs saved comment threads.
Not a fit when
- You lack rights to download or reuse the material.
- The video is outside your authorized access.
- Subtitles are expected in every language regardless of source availability.
Example request
Fetch only metadata and available subtitles for this authorized video, without downloading video or audio.MCP
- Transport
- stdio
- Authentication
- not required
Limitations
Install yt-dlp separately; the MCP npm package does not replace the system downloader. Video, subtitle and comment availability depends on the platform and source, so inspect metadata first. Downloads write files and consume storage; set resolution and segment limits in advance. If configured to use cookies, the tool receives the associated site session; protect that data and keep it out of chat and repositories.
How to disable. Remove yt-dlp from the MCP client and stop active downloads. Downloaded files and configured cookies remain for separate management.
FAQ
Can I read metadata without downloading?
Yes. Metadata tools do not require a media download.
Which subtitle formats are available?
It downloads VTT and can produce clean transcript text.
Where do downloads go by default?
The user’s Downloads directory, configurable as needed.
Can I download audio only?
Yes. Dedicated audio operations use supported output formats.
Why is a local URL rejected?
Local and private-network addresses are blocked by default through URL validation.
Discussion
No questions or reviews yet.
Related
MCP serversyt-dlp is a media-download CLI that lists available formats and simulates a download before saving media files to disk.
MCP server for YouTube transcripts: an agent gets a video transcript by URL, with timestamps and metadata
MCP server that gives an agent a YouTube video's transcript from its link, with language choice, timestamps and ads filtered out
Cloudglue MCP: video analysis for agents
Cloudglue MCP Server
An agent in Cursor or Claude Desktop describes video, splits it into chapters and shots and finds moments by speech or picture
YouTube MCP (Zubeid)
Zubeid YouTube MCP Server
YouTube MCP on a Data API key: an agent searches videos and channels with a subscriber-count filter and reads playlists, statistics and transcripts
Claude Video Vision
claude-video-vision
An MCP server and Claude Code plugin that let the agent watch video: frames via ffmpeg and audio transcription via Gemini, local Whisper or OpenAI
README in short
Response-length limits bound agent context use. Comment tools can produce Markdown discussion trees.