YouTube Research MCP

YouTube Research MCP

[CI] [npm] [npm downloads] [GitHub stars] [LightNow MCP capabilities] [MIT License]

Turn YouTube videos into citation-ready research for Codex, Claude, Cursor, and other MCP clients. Paste a video URL and get transcript evidence with timestamps and links that open at the exact quoted moment.

[YouTube Research MCP turns a video URL into focused timestamp-linked evidence]

Input:  https://youtu.be/dQw4w9WgXcQOutput: [01:05] ...evidence text...        https://www.youtube.com/watch?v=dQw4w9WgXcQ&t=65s

No YouTube API key is required for transcript research.

Why this server

• Citation-ready research — every transcript segment includes a timestamp and source URL.
• Zero-key quick start — transcripts work immediately; add an API key only for search and analytics.
• URL-first input — accepts normal, short, Shorts, embed, live, and raw video-ID formats.
• Agent-efficient filtering — search inside transcripts and return nearby context instead of spending tokens on the entire video.
• Full YouTube intelligence — optional API mode adds comments, video/channel statistics, trends, and comparisons.
• Remote-native MCP — Streamable HTTP transport at /mcp, plus Docker and Smithery support.

Architecture and trust boundary

flowchart LR    C[MCP client] -->|stdio or Streamable HTTP| T[Transport]    T --> R[Tool and resource registry]    R --> S[YouTubeService]    S --> K[(In-memory cache)]    S -->|no key required| YT[Caption tracks]    S -->|optional API key| API[YouTube Data API v3]

The MCP layer validates inputs and returns structured evidence; it does not ask a model to invent summaries or citations. Transcript tools derive timestamps and canonical links from caption segments. Data API tools return an explicit key-required error when YOUTUBE_API_KEY is unavailable, and remote authentication is enforced before MCP session handling.

Focused evidence, not a transcript dump

The included live smoke test asks a focused question about a public video and compares the response with the full timestamped transcript:

This measures response characters, not model-specific tokens. Reproduce it against the default public fixture—or substitute your own video and query:

npm run test:liveLIVE_TEST_VIDEO='https://youtu.be/VIDEO_ID' LIVE_TEST_QUERY='evaluation' npm run test:live

Quick start — no API key

Requirements: Node.js 22 or newer.

The shortest local install uses stdio and needs no API key:

codex mcp add youtube-research -- npx -y @coyasong/youtube-mcp-server@latest

For Claude Desktop and other MCPB-compatible desktop clients, download the latest one-click MCP bundle and open it. The bundle vendors its runtime dependencies, starts locally over stdio, and asks for a YouTube API key only if you want the optional analytics tools.

[Install in VS Code]

Or install from the VS Code command line:

code --add-mcp '{"name":"youtube-research","command":"npx","args":["-y","@coyasong/youtube-mcp-server@latest"]}'

For Claude Desktop, Cursor, and other stdio clients, use command npx with arguments -y @coyasong/youtube-mcp-server@latest.

To run the Streamable HTTP transport instead:

git clone https://github.com/coyaSONG/youtube-mcp-server.gitcd youtube-mcp-servernpm installnpm run buildnpm start

The server starts at http://localhost:3000/mcp in transcript-only mode. Confirm it with:

curl http://localhost:3000/health

Connect from Codex

With the HTTP server running:

codex mcp add youtube-research --url http://localhost:3000/mcp

Then ask Codex:

Use research-video to find what this video says about evaluation,and cite the exact moments: https://www.youtube.com/watch?v=VIDEO_ID

Connect with MCP Inspector

npx @modelcontextprotocol/inspector# Connect to http://localhost:3000/mcp

Install through Smithery

npx -y smithery@latest auth loginnpx -y smithery@latest mcp add coyaSONG/youtube-mcp-server --client claude

Enable search, comments, and analytics

Set a YouTube Data API v3 key to switch from transcript-only to full mode:

export YOUTUBE_API_KEY=your_key_herenpm start

YOUTUBE_API_KEY is optional. PORT defaults to 3000. For stdio clients, set YOUTUBE_API_KEY in the MCP server environment configuration.

Secure a remote deployment

Do not expose a full-mode server publicly without authentication: unauthenticated users could consume your YouTube API quota. Set a strong bearer token and restrict browser origins when deploying outside localhost:

export MCP_BEARER_TOKEN='replace-with-a-long-random-secret'export CORS_ORIGIN='https://your-client.example'export MAX_SESSIONS=100export SESSION_IDLE_TIMEOUT_MS=1800000npm start

Browser origins are denied by default. CORS_ORIGIN accepts a comma-separated allowlist of exact http:// or https:// origins; wildcard origins are rejected.

Connect Codex using an environment variable rather than writing the secret into its configuration:

export YOUTUBE_MCP_TOKEN='replace-with-a-long-random-secret'codex mcp add youtube-research \  --url https://your-server.example/mcp \  --bearer-token-env-var YOUTUBE_MCP_TOKEN

/health remains public for container health checks. MCP requests return 401 when authentication is enabled and the bearer token is missing or invalid. Idle sessions are removed automatically, and MAX_SESSIONS bounds memory use.

Best first tool

research-video accepts:

• video: a YouTube URL or 11-character video ID
• language: optional caption language such as en, ko, or ja
• query: optional phrase to find inside the transcript
• contextLines: surrounding segments to retain, from 0 to 5
• matchMode: word (default) or substring
• startSeconds / endSeconds: optional time window
• offset: result offset for pagination
• maxSegments: response cap from 1 to 1,000 (default: 200)

It returns structured JSON containing the video title and channel identity, canonical video URL, full caption-track duration and segment count, matching transcript segments, human-readable citation labels, timestamps, directly navigable citation URLs, and pagination metadata. For long videos, use a query or time window first; follow nextOffset only when more evidence is needed.

Compare evidence across videos

research-videos applies one focused query to 2–5 video URLs concurrently. It returns the same structured, timestamp-linked evidence for each source while capping results per video. This is useful for comparing interviews, checking whether multiple sources support a claim, or researching a topic across a short watchlist.

{  "videos": [    "https://youtu.be/VIDEO_ONE",    "https://youtu.be/VIDEO_TWO"  ],  "query": "evaluation",  "maxSegmentsPerVideo": 10}

See YouTube Research Recipes for copy-paste workflows for fact-checking claims, comparing interviews, navigating long talks, researching multilingual captions, and building citation-ready notes.

Capability modes

Captions must be available for the requested video. Age-restricted, private, region-restricted, or caption-disabled videos may not return a transcript.

If this project saves you research time, consider starring the repository so other agent builders can discover it.

See the broader AI tooling portfolio for the reliability principles shared with ralph-research and tmuxicate.

Docker

docker build -t youtube-research-mcp .docker run --rm -p 3000:3000 youtube-research-mcp
# Full modedocker run --rm -p 3000:3000 -e YOUTUBE_API_KEY=your_key youtube-research-mcp

Development

npm run dev             # HTTP server from TypeScriptnpm test                # build and run all testsnpm run test:live       # live public-video transcript and citation smoke testnpm run test:user       # clean-room smoke test against the published npm packagenpm run test:mcpb       # build, unpack, and exercise the installable MCP bundle

Maintainers can follow the release guide. Published GitHub releases run the complete npm, MCP Registry, MCPB, and Smithery delivery pipeline.

API Reference

Resources

• youtube://video/{videoId} - Get detailed information about a specific video
• youtube://channel/{channelId} - Get information about a specific channel
• youtube://transcript/{videoId} - Get transcript for a specific video
  • Optional query parameter: ?language=LANGUAGE_CODE (e.g., en, ko, ja)

Tools

Basic Tools

• research-video - Get citation-ready transcript evidence from a URL or video ID without an API key
• research-videos - Compare timestamp-linked evidence across 2–5 videos without an API key
• search-videos - Search for YouTube videos with advanced filtering options
• get-video-comments - Get comments for a specific video
• get-video-transcript - Get transcript for a specific video with optional language
• enhanced-transcript - Advanced transcript extraction with filtering, search, and multi-video capabilities
• get-key-moments - Extract key moments with timestamps from a video transcript for easier navigation
• get-segmented-transcript - Divide a video transcript into segments for easier analysis

Tools requiring YOUTUBE_API_KEY are search, comments, statistics, discovery, and channel analysis. enhanced-transcript needs the key only when includeMetadata is true.

Statistical Tools

• get-video-stats - Get statistical information for a specific video
• get-channel-stats - Get subscriber count, view count, and other channel statistics
• compare-videos - Compare statistics across multiple videos

Discovery Tools

• get-trending-videos - Retrieve trending videos by region and category
• get-video-categories - Get available video categories for a specific region

Analysis Tools

• analyze-channel-videos - Analyze performance trends of videos from a specific channel

Prompts

• video-analysis - Generate an analysis of a YouTube video
• transcript-summary - Generate a summary of a video based on its transcript with customizable length and keywords extraction
• segment-by-segment-analysis - Provide detailed breakdown of content by analyzing each segment of the video

Examples

Accessing a Video Transcript

youtube://transcript/dQw4w9WgXcQ

Getting a Transcript in a Specific Language

youtube://transcript/dQw4w9WgXcQ?language=en

Using the Statistical Tools

// Get video statistics{  "type": "tool",  "name": "get-video-stats",  "parameters": {    "videoId": "dQw4w9WgXcQ"  }}
// Compare multiple videos{  "type": "tool",  "name": "compare-videos",  "parameters": {    "videoIds": ["dQw4w9WgXcQ", "9bZkp7q19f0"]  }}

Using the Transcript Summary Prompt

{  "type": "prompt",  "name": "transcript-summary",  "parameters": {    "videoId": "dQw4w9WgXcQ",    "language": "en"  }}

Using the Enhanced Transcript Tool

// Basic multi-video transcript extraction{  "type": "tool",  "name": "enhanced-transcript",  "parameters": {    "videoIds": ["dQw4w9WgXcQ", "9bZkp7q19f0"],    "format": "timestamped"  }}
// With search and time filtering{  "type": "tool",  "name": "enhanced-transcript",  "parameters": {    "videoIds": ["dQw4w9WgXcQ"],    "filters": {      "timeRange": {        "start": 60,  // Start at 60 seconds        "end": 180    // End at 180 seconds      },      "search": {        "query": "never gonna",        "contextLines": 2      }    },    "format": "merged"  }}
// With duration-based segmentation for easier analysis{  "type": "tool",  "name": "enhanced-transcript",  "parameters": {    "videoIds": ["dQw4w9WgXcQ"],    "filters": {      "segment": {        "count": 5,        "method": "smart"  // Balances caption duration across segments      }    },    "format": "timestamped",    "language": "en"  }}

Using the Enhanced Transcript Analysis Features

// Get key moments from a video{  "type": "tool",  "name": "get-key-moments",  "parameters": {    "videoId": "dQw4w9WgXcQ",    "maxMoments": 5  }}
// Get a segmented transcript{  "type": "tool",  "name": "get-segmented-transcript",  "parameters": {    "videoId": "dQw4w9WgXcQ",    "segmentCount": 4  }}
// Get a segment-by-segment analysis{  "type": "prompt",  "name": "segment-by-segment-analysis",  "parameters": {    "videoId": "dQw4w9WgXcQ",    "segmentCount": 4  }}
// Get customized transcript summary{  "type": "prompt",  "name": "transcript-summary",  "parameters": {    "videoId": "dQw4w9WgXcQ",    "language": "en",    "summaryLength": "detailed",    "includeKeywords": true  }}

Error Handling

The server handles various error conditions, including:

• Invalid or missing API key for Data API tools
• Video or channel not found
• Transcript not available
• Network issues

License

MIT

Acknowledgements

• Model Context Protocol TypeScript SDK
• YouTube Data API
• YouTube.js

Comment pagination and evidence limits

search-videos accepts the provider pageToken. get-video-comments preserves the next top-level token and accepts searchTerms for YouTube-side filtering.

Use includeReplies: true for the embedded reply subset, or completeReplies: true for bounded reply pagination. Full retrieval defaults to 100 replies per thread; maxRepliesPerThread accepts 1–1000. Across all threads, each call returns at most 1000 replies and 1 MiB of reply JSON and makes at most 20 additional reply requests. Each Data API request times out after 10 seconds and does not retry automatically.

Inspect each thread's replyCoverage: available is the provider's reported count, returned counts unique replies, complete confirms full observed coverage, and truncated indicates a local budget or pagination guard stopped collection. Counts may change while YouTube comments are being edited. Missing IDs or early provider termination do not establish complete coverage. Reduce the top-level maxResults to reserve more of the shared budget for each thread; pageToken continues top-level threads, not an incomplete reply list.

Research source metadata includes captionRequestedLanguage, captionTrackLanguage, captionLanguage, captionOrigin (manual, youtube_asr, or translated) and captionSourceOrigin. Unknown provenance is null.