npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

youtube-relay-mcp

v1.3.0

Published

Code-execution-based YouTube tool — CLI (ytrelay) + MCP shim + Claude Code skill. Search YouTube, fetch transcript/title/description, get an embeddable video ID.

Downloads

326

Readme

youtube-relay-mcp

A deep-research tool for YouTube for AI agents — a TypeScript CLI (ytrelay), an MCP server (youtube-relay-mcp), and a Claude Code skill. Cast a wide net, rank candidates cheaply on metadata, peek before committing, and read full transcripts only for the finalists. One engine: youtubei.js.

Status: all four commands work — enriched, captioned-only-by-default search (views, recency, verified, snippet); rich info (full description, chapters, caption availability); transcripts via the signed timedtext caption URL (no PO token / external binary); a peek tier (--head/--max-chars); and batch + stdin ergonomics. The bundled SKILL.md teaches the recommended funnel.

Install

npm i -g youtube-relay-mcp

CLI

Every command prints a JSON envelope to stdout.

ytrelay search "agentic engineering" --limit 30 --sort views   # cheap, enriched, captioned-only
ytrelay info <id|url> [<id> ...]      # full description + chapters + caption info (batch ok)
ytrelay transcript <id|url> --head 120   # PEEK: first 2 min; or --max-chars N
ytrelay context <id|url>              # metadata + full transcript + embed in one shot
ytrelay frame <id|url> --at 1:30      # high-res still frame at a timestamp (needs ffmpeg + yt-dlp)

frame extracts full-resolution stills at exact timestamps (pairs with transcript startMs to see a moment). It's the only command needing external binaries (ffmpeg + yt-dlp on PATH); the rest are pure-Node.

The funnel (see SKILL.md): rank on cheap search metadata → enrich the shortlist with info → peek finalists with transcript --head → full read only the survivors. --features all widens search beyond captioned videos. Every result carries embedUrl (https://www.youtube.com/embed/<id>).

Why a CLI + skill (not a standalone MCP server)

The task is a stateless search/fetch pipeline. An agent shells out to the CLI, taught by the bundled SKILL.md. The MCP shim is provided for parity and non-CLI hosts. See PLAN.md for the full design rationale.

Development

bun install
bun run check     # typecheck + lint + test
bun test --watch  # TDD
bun run build     # → dist/

License

MIT © Tamas Gabor