Vidlens Mcp
by thatsrajancommunityv1.2.9
23
3
VidLens is a Model Context Protocol server that enables AI agents to search YouTube, read transcripts, and synthesize information across multiple videos, providing detailed insights and analysis without requiring an API key.
From the registry: YouTube as a queryable database for AI agents. 41 tools, zero config.
Install & configure
Prerequisites
- required
- requiredyt-dlpAuto-installed during setup, or manually via setup command
$ npx vidlens-mcp setup - optionalffmpegOptional - needed for frame extraction and visual indexing
$ brew install ffmpeg
Install in your client
Paste into the AI you're chatting with — it will figure out the rest
Please install the `vidlens-mcp-thatsrajan` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Node.js 22+** — Required - uses node:sqlite built-in (https://nodejs.org/)
- **yt-dlp** — Auto-installed during setup, or manually via setup command Run: `npx vidlens-mcp setup`
Optional prerequisites:
- ffmpeg — Optional - needed for frame extraction and visual indexing Run: `brew install ffmpeg`
Canonical MCP server config (stdio transport):
- command: `npx`
- args: ["-y","vidlens-mcp","serve"]
- optional environment variables:
- `YOUTUBE_API_KEY`: YouTube Data API v3 key for better metadata and comments (example: `<your-youtube-api-key>`)
- `GEMINI_API_KEY`: Google Gemini API key for higher-quality embeddings and visual analysis (example: `<your-gemini-api-key>`)
Note: YouTube intelligence MCP with 41 tools across 10 modules. Works without API keys via fallback chain. Supports semantic search, visual search (macOS Apple Vision), sentiment analysis, trend discovery. Run 'npx vidlens-mcp setup' for auto-configuration.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.Note: YouTube intelligence MCP with 41 tools across 10 modules. Works without API keys via fallback chain. Supports semantic search, visual search (macOS Apple Vision), sentiment analysis, trend discovery. Run 'npx vidlens-mcp setup' for auto-configuration.
Tools (10)
exploreYouTube
Intent-aware search with multi-query ranking, parallel enrichment, transcript summaries, structured benchmark data, and background indexing.
findVideos
Search YouTube by query with metadata.
inspectVideo
Deep metadata - tags, engagement, language, category.
inspectChannel
Channel stats, description, recent uploads.
readTranscript
Full transcript with timestamps and chapters.
readComments
Top comments with likes and engagement.
importPlaylist
Index an entire playlist's transcripts.
measureAudienceSentiment
Comment sentiment with themes and risk signals.
downloadAsset
Download video, audio, or thumbnails.
indexVisualContent
Extract frames, run Apple Vision OCR + feature prints, Gemini frame descriptions, and Gemini semantic embeddings.
Demos
Packages
vidlens-mcp
npmv1.2.9transport: stdio
Details
- Language
- TypeScript
- License
- MIT
- Released
- Apr 8, 2026
- Last commit
- Apr 8, 2026
- Registry name
- io.github.thatsrajan/vidlens-mcp