The YouTube MCP server acts as a bridge between video content and AI assistants like Claude Desktop, allowing users to interact with the world's largest video platform directly through their chat interface. In its simplest form, it lets an AI "watch" and "listen" to YouTube videos by searching for relevant topics and pulling in text-based transcripts. This means users can ask their AI to find tutorials, summarize speeches, or explain complex concepts from specific videos without ever having to manually copy and paste text. Under the hood, this tool provides specialized functions including `search_youtube_videos` and `get_youtube_transcript`. It can search for up to 50 videos based on a specific query and extract full transcripts complete with timestamps and metadata. For more complex workflows, the server includes a pre-built analysis prompt template designed to help Large Language Models (LLMs) perform comprehensive content reviews or comparative studies across multiple videos simultaneously. For developers and AI power users, this server is built on the `fastmcp` framework and utilizes the YouTube Data API v3 for robust data retrieval. It is designed for seamless integration into the Claude Desktop configuration, requiring only a Google Cloud API key to get started. By automating the retrieval of structured video data and text, it empowers LLMs to perform sophisticated tasks like cross-referencing academic lectures, conducting deep market research, or generating automated summaries of video trends, making it an essential tool for building video-aware AI agents.
Category: Design, Media & Creative
Tags: media, search, transcripts, video, youtube
https://github.com/bendelpino/youtube-mcp). 2. Create a virtual environment: bash python3 -m venv .venv 3. Activate the virtual environment: bash source .venv/bin/activate 4. Install dependencies: bash pip install -r requirements.txt 5. Obtain a YouTube API Key: * Go to the Google Cloud Console. * Enable the YouTube Data API v3. * Create credentials to obtain an API Key..env file in the project root: bash # Copy example environment file cp .env.example .env # Edit .env and add your YouTube API key YOUTUBE_API_KEY=your_actual_api_key_herejson { "mcpServers": { "youtube": { "command": "local/path/to/uv", "args": [ "run", "--directory", "/Path/to/your/project", "youtube_server.py" ], "env": { "YOUTUBE_API_KEY": "your_key_here" } } } }search_youtube_videos: Search YouTube for videos based on a query. * Parameters: search_term (string), num_videos (int, default: 5). * get_youtube_transcript: Extract transcripts from a YouTube video using a URL or video ID. * Parameters: video_url_or_title (string). * analyze_youtube_content_prompt: AI prompt template for comprehensive YouTube content analysis. * Parameters: search_term (string), num_videos (int).get_youtube_transcript to fetch the text and then provides a bulleted list of the three most important API changes mentioned in the video.search_youtube_videos and analyze_youtube_content_prompt tools, the AI can perform a bulk analysis of the current landscape. It can search for a topic, retrieve metadata for the top results, and pull transcripts to identify recurring keywords, sentiment, or gaps in the conversation. Example: A product manager asks, "Search for the top 5 videos about 'sustainable fashion' from the last month and tell me what specific materials are being mentioned most frequently." The MCP searches, fetches transcripts, and the AI synthesizes the data.get_youtube_transcript, the AI can act as a professional editor to restructure the raw speech into polished, written content while maintaining the creator's original insights. Example: A creator gives the AI a link to their latest tutorial and says, "Get the transcript for this video and rewrite it into a step-by-step 'How-To' blog post with a catchy title and a 150-word introduction."Part of MCP Servers