YouTube IntelligenceSAFE
MCP server for YouTube transcript extraction, channel monitoring, and content intelligence
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
[](https://www.python.org/) [](LICENSE) [](https://modelcontextprotocol.io/) [](https://pypi.org/project/mcp-youtube-intelligence/)
🌐 English | 한국어
YouTube 영상을 지능적으로 분석하는 [MCP](https://modelcontextprotocol.io/) 서버 + CLI MCP (Model Context Protocol)는 Claude, Cursor 같은 AI 도구가 외부 서비스를 사용할 수 있게 해주는 표준 프로토콜입니다. 이 서버를 연결하면 "이 영상 요약해줘" 한마디로 분석이 완료됩니다.
🎯 핵심 가치: 원본 자막(2,000~30,000 토큰)을 서버에서 처리하여 LLM에는 ~200–500 토큰만 전달합니다.
🤔 왜 이 서버인가?
대부분의 YouTube MCP 서버는 원본 자막을 그대로 LLM에 던집니다.
🚀 빠른 시작
1. 설치
pip install mcp-youtube-intelligence pip install yt-dlp # 자막 추출에 필요
💡 LLM 없이도 기본 요약(핵심 문장 추출)은 동작합니다. 고품질 요약을 원하면 아래 LLM 설정을 참고하세요.
2. 첫 번째 명령어 실행
# 리포트 생성 — 요약, 토픽, 엔티티, 댓글을 한번에 분석 (LLM 연동필요) mcp-yt report "https://www.youtube.com/watch?v=LV6Juz0xcrY" # 자막 요약만 mcp-yt transcript "https://www.youtube.com/watch?v=LV6Juz0xcrY" # 영상 ID만 써도 됩니다 mcp-yt report LV6Juz0xcrY
⚠️ zsh 사용자: URL에 ?가 있으므로 반드시 따옴표로 감싸세요.📋 리포트 출력 예시
mcp-yt report "https://www.youtube.com/watch?v=LV6Juz0xcrY" 실행 결과 (extractive 요약):
# 📹 Video Analysis Report: OpenClaw Use Cases that are Actually Helpful! (ClawdBot) > Channel: Duncan Rogoff | AI Automation | Duration: 16:29 | Language: en_ytdlp ## 1. Summary OpenClaw is the most powerful AI agent frame
306b2c0500daOBSERVED · 2026-10-08Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add mcp-youtube-intelligence --env ANTHROPIC_API_KEY=${ANTHROPIC_API_KEY} --env GOOGLE_API_KEY=${GOOGLE_API_KEY} --env MYI_YOUTUBE_API_KEY=${MYI_YOUTUBE_API_KEY} --env OPENAI_API_KEY=${OPENAI_API_KEY} -- uvx mcp-youtube-intelligence{
"mcpServers": {
"mcp-youtube-intelligence": {
"command": "uvx",
"args": [
"mcp-youtube-intelligence"
],
"env": {
"ANTHROPIC_API_KEY": "${ANTHROPIC_API_KEY}",
"GOOGLE_API_KEY": "${GOOGLE_API_KEY}",
"MYI_YOUTUBE_API_KEY": "${MYI_YOUTUBE_API_KEY}",
"OPENAI_API_KEY": "${OPENAI_API_KEY}"
}
}
}
}Exposed tools (10)
10 read · 0 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
extract_entities | read | Extract structured entities (companies, indices, people, sectors, etc.) from a video transcript. |
generate_report | read | Generate a structured markdown report for a YouTube video. Includes summary, topic segments, entities, and optionally comments. |
get_comments | read | Get top comments for a video. Optionally summarize them. |
get_playlist | read | Get playlist metadata and video list from a YouTube playlist. |
get_transcript | read | Get video transcript. mode: |
get_video | read | Get video metadata + summary (~300 tokens). Provide a YouTube video ID. |
monitor_channel | read | Monitor a YouTube channel via RSS. action: |
search_transcripts | read | Search stored transcripts by keyword. Returns matching snippets. |
search_youtube | read | Search YouTube videos by keyword. Returns metadata list (~200 tokens). |
segment_topics | read | Segment a video transcript into topics based on transition markers. |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- none-observed
- Network
- declared (7 observation(s))
- Shell
- none-observed
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (0)
No findings outside the package's declared scope.
Gates applied: no_behavioural_pass.
306b2c0500dafull audit observations/trust-audit/mcp-server/janghyuckyun__youtube-intelligence.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-08 | 306b2c0500da | SAFE | B | 89 | first audit |
Questions
What is the YouTube Intelligence MCP server?
MCP server for YouTube transcript extraction, channel monitoring, and content intelligence
What tools does YouTube Intelligence expose?
10 in total: 10 read-only, 0 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is YouTube Intelligence safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B.
What credentials does YouTube Intelligence need?
It reads ANTHROPIC_API_KEY, GOOGLE_API_KEY, MYI_YOUTUBE_API_KEY and OPENAI_API_KEY from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How does YouTube Intelligence run?
It speaks stdio, so it runs as a local process your client starts. It is published on PyPI as mcp-youtube-intelligence.
How current is this page?
The grade is for one exact copy of the source (306b2c0500da), read on 2026-10-08. The repository is watched and re-audited when it changes.