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

video-action-mcp

v0.1.0

Published

MCP server that lets Claude watch and listen to videos (local files or YouTube links) through Video Action Agent: transcript, on-screen text, and what was observed vs. not confirmed.

Readme

video-action-mcp

Servidor MCP que le da a Claude (Claude Code, Claude Desktop) ojos y oídos para videos a través de Video Action Agent: imagen, audio, voz y texto en pantalla. Cada resultado separa lo observado en el video, lo no confirmado y las sugerencias de la IA.

English below.

Herramientas

| Herramienta | Qué hace | |---|---| | analyze_video_file(path) | Sube un video de tu computadora (.mp4 .mov .m4v .webm .3gp, hasta 500 MB y 30 min) y devuelve el análisis | | analyze_youtube(url) | Analiza un video público de YouTube (no se descarga nada) | | get_analysis(jobId) | Estado o resultado de un análisis (también de los hechos en la app) |

Opcionales en las tres: lang (es/en), wait_seconds (cuánto esperar; si no termina, devuelve el jobId), format (markdown o json). En las dos primeras, mode (auto por defecto; compare y max_precision usan además OpenAI).

El resultado trae la transcripción completa, todo el texto en pantalla, comandos y código mostrados (marca los peligrosos, como curl … | bash) y el texto del video que intenta darle órdenes a una IA, que llega marcado como dato.

Instalar

  1. En la app: menú ⋮ → Conectar con Claude → Crear llave. Te da el comando ya armado con tu llave; la llave no se vuelve a mostrar.
  2. Pégalo en tu terminal:
claude mcp add video-action --scope user -e VAA_API_KEY=vaa_… -- npx -y video-action-mcp

En Windows, con cmd /c delante de npx:

claude mcp add video-action --scope user -e VAA_API_KEY=vaa_… -- cmd /c npx -y video-action-mcp

Desde el código fuente (sin npm): npm install dentro de mcp/ y usa -- node /ruta/a/VideoActionAgent/mcp/index.js en lugar de npx.

Claude Desktop: en claude_desktop_config.json:

{
  "mcpServers": {
    "video-action": {
      "command": "npx",
      "args": ["-y", "video-action-mcp"],
      "env": { "VAA_API_KEY": "vaa_…" }
    }
  }
}

Variables de entorno

| Variable | | |---|---| | VAA_API_KEY | Tu llave personal (vaa_…). Obligatoria | | VAA_API_URL | Servidor. Por defecto https://videoactionagent.web.app | | VAA_LANG | Idioma por defecto de las respuestas: es (por defecto) o en |

La llave

  • Actúa como tu cuenta: tus análisis aparecen en tu historial de la app y cuentan contra tus límites diarios (también si tu cuenta es de administrador).
  • Solo sirve para analizar videos y leer esos análisis. Ver tu historial, crear llaves o el panel del dueño exigen la sesión de la app.
  • El servidor guarda solo su hash. Si se filtra, revócala en la misma pantalla y crea otra (en hasta un minuto deja de valer en todas partes).
  • Máximo 5 por cuenta.

Licencia

MIT, solo para este conector (los archivos de este paquete). El servicio Video Action Agent —su servidor, el análisis y la app— no forma parte del paquete y no tiene licencia abierta.


English

MCP server that gives Claude eyes and ears for videos through Video Action Agent: analyze_video_file(path), analyze_youtube(url) and get_analysis(jobId). Each result includes the full transcript, on-screen text, commands and code shown, and keeps observed / not confirmed / AI suggestions apart.

Create a personal key in the app (⋮ → Connect to Claude), then:

claude mcp add video-action --scope user -e VAA_API_KEY=vaa_… -- npx -y video-action-mcp

(On Windows put cmd /c before npx.) Set VAA_LANG=en for English answers. The key acts as your account (your history, your daily limits) but can only analyze videos and read those analyses; revoke it from the same screen.

License: MIT, for this connector only. The Video Action Agent service (its server, the analysis and the app) is not part of this package and is not open source.