# VidWords YouTube [Health: Active]

**Category:** 💻 Developer Tools  
**Repository:** https://vidwords.com/resources/youtube-mcp-server  
**Views:** 0  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/vidwords-youtube

## Description
Search YouTube transcripts and read a video's frames; answers cite clickable timestamps.

## Claude Desktop Quick Installation
No confirmed install details yet, so no config is published. Follow the setup steps in the repository above.

## Documentation

## What VidWords YouTube MCP server does

VidWords YouTube MCP server gives AI agents access to two YouTube-focused tasks: searching video transcripts and reading a video's frames. This supports questions that depend on spoken content as well as requests that require inspecting what appears in the video itself.

The result format includes clickable timestamps. A developer can use those links to move from an answer to the corresponding location in the source video, which helps with checking context and locating evidence. The available description does not specify whether transcript search applies to one supplied video, a broader set of videos, or both, so the exact search scope should be confirmed before implementation.

## Tools and capabilities

The documented capabilities are:

- Search YouTube transcripts.
- Read frames from a YouTube video.
- Return answers with clickable timestamp citations.

Together, these capabilities cover both text-based and visual video retrieval. Transcript search is relevant when the needed information is spoken or appears in captions. Frame reading is relevant when the answer depends on visible content rather than words alone. Timestamp links provide a way to review the cited moment instead of relying only on generated text.

VidWords YouTube MCP server is therefore a candidate for agents that need to analyze video material as a source. It may fit workflows where a user asks for information from a video and expects references that lead back to the source timeline. The supplied material does not document additional YouTube operations such as downloading media, publishing videos, managing channels, or accessing comments, so those functions should not be assumed.

## Limitations and notes

The source information is limited to the server's stated YouTube capabilities. It does not identify the MCP transport, installation command, package format, authentication method, required environment variables, supported models, or supported MCP clients. No license information is provided either.

The description also does not explain how frame selection works, whether frames are returned as images or another representation, how transcript search is queried, or how timestamp citations are formatted beyond being clickable. These details matter when designing an agent workflow and should be verified from the project's current documentation before deployment.

Use the server when transcript retrieval, frame inspection, and timestamp-linked answers are the core requirements. Do not treat the listing as evidence that it can perform broader YouTube account or media-management tasks. Its documented role is limited to finding transcript content, reading video frames, and citing locations in the video.

_Full upstream README: https://allmcps.com/mcp/vidwords-youtube/readme_