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

@arpproject/rocrate-mcp-server

v1.0.4

Published

ARP RO-Crate MCP Server exposing RO-Crate editing and profile-aware validation tools for use in ARP

Readme

ARP RO-Crate MCP Server

@arpproject/rocrate-mcp-server is an MCP server for RO-Crate editing, validation, and profile-aware constraints following the best practices of the ARP project (https://researchdata.hu/). It is designed to be used with MCP-compatible AI assistants such as Codex, Claude Code. It can be used with the schemas and profiles offered by the ARP Schema Registry (https://cedar.schema.researchdata.hu/)

Copyright 2026, SZTAKI DSD, (https://dsd.sztaki.hu/). Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0

Quick start · Első lépések

Quick start

@arpproject/rocrate-mcp-server makes it easy to create RO-Crate packages using MCP-compatible AI assistants and subsequently upload them to the ARP system.

To use it, you need an MCP-compatible AI assistant, such as Codex, Claude Code, OpenCode, Kilo Code, Roo Code, Gemini CLI, or Qwen Code.

Node.js 18 or newer is required for installation:

npm install -g @arpproject/rocrate-mcp-server

Note: when upgrading an existing installation, the package first asks the shared RO-Crate MCP daemon to shut down gracefully through its socket (or Windows named pipe). Restart any AI assistant that already has an MCP connection after the upgrade.

To configure it for a particular AI assistant, use the interactive installer:

rocrate-mcp-server -i

Select the AI assistant you want to use, and the installer will automatically create the required configuration.

The selection screen supports the up/down arrow keys as well as entering an agent number. Before asking for confirmation, the installer displays the exact MCP section it will write and the destination file. Press Enter to accept, or use an explicit agent ID when preferred:

rocrate-mcp-server -i codex

Supported IDs are codex, claude, opencode, kilo, roo, gemini, and qwen. The generated configuration launches the standalone server through the shared per-user RockIT socket (~/.rockit/rocrate-mcp-server.sock).

The standalone server and RocKIT share the same per-user profile and schema storage by default: ~/.rockit. Set ROCKIT_ROOT_PATH to use another shared root, and use the ROCKIT_* profile environment variables for filename and provider-storage overrides.

You can verify that the installation was successful by starting the AI assistant and listing the configured MCP servers. This is typically done using the /mcp command (for example, in Codex and Claude Code) or /mcps (for example, in OpenCode). The list should contain an MCP server named rocrate that runs the rocrate-mcp-server command.

To use rocrate-mcp-server, start your AI assistant in the directory where you want to create or edit the RO-Crate. Give the assistant the appropriate instructions for creating the RO-Crate package, and it will automatically start using rocrate-mcp-server and follow the workflows provided by the server.If RO-Crate editing is not using the MCP server automatically, you can explcitily ask the assistant to "Use the RO-Crate MCP server" to start using it.

At the end of the workflow, the completed RO-Crate package (the ro-crate-metadata.json file) can be opened in the ARP AROMA software. The assistant will usually offer to do this automatically. If it does not, simply ask it to

 "Open the dataset in AROMA".

The assistant will provide a URL that opens AROMA in a browser with the RO-Crate package you are currently editing, or in case of a more intelligent agent, it may even open the browser automatically.

If you place the AI assistant and the AROMA browser window side by side, changes made to the RO-Crate through the assistant will immediately appear in AROMA, where they can be reviewed and also edited directly. It is also to a good experience to work in a desktop app, like ChatGPT dekstop, and have the AROMA browser window open inside the app. The built-in browser can also be used to perform tasks, whichrequire interactive browser access, ie. "browser use"

The server also provides a local dashboard at http://127.0.0.1:9393 for monitoring MCP activity and configure various values.

rocrate-mcp-server can also upload the dataset to ARP Dataverse; simply ask your assistant to do so. For uploads to work, set the DATAVERSE_API_KEY environment variable. You can get it from https://repo.researchdata.hu/dataverseuser.xhtml?selectTab=apiTokenTab. The assistant will also ask for this value if they have not been configured.

The dashboard Settings page lets you enter or override the TAVILY_API_KEY, DATAVERSE_BASE_URL, and DATAVERSE_API_KEY values for the running MCP process, and controls whether successful Dataverse uploads keep their generated RO-Crate ZIP files. Environment variables remain the defaults; dashboard settings are saved in ~/.rockit/rocrate-mcp-settings.json. API-key overrides can be cleared from Settings to restore the environment fallback. The file contains configured API keys in plaintext and uses user-only permissions where supported.

Első lépések

Az @arpproject/rocrate-mcp-server segítségével MCP-kompatibilis AI-asszisztenseket használva egyszerűen hozhatók létre RO-Crate csomagok, amelyeket aztán az ARP rendszerébe is fel lehet tölteni.

A használatához szükség van egy MCP-kompatibilis AI-asszisztensre, például Codex, Claude Code, OpenCode, Kilo Code, Roo Code, Gemini CLI vagy Qwen Code.

A telepítéshez Node.js 18 vagy újabb szükséges:

npm install -g @arpproject/rocrate-mcp-server

Megjegyzés: meglévő telepítés frissítésekor a csomag először szabályosan leállítja a közös RO-Crate MCP démont a socketen (Windowson named pipe-on) keresztül. A frissítés után indítsd újra azt az AI-asszisztenst, amely már az új szervert fogja elindítani és használni.

Egy adott AI-asszisztenshez való konfiguráláshoz használd az interaktív telepítőt:

rocrate-mcp-server -i

Itt válaszd ki a használni kívánt AI-asszisztenst, és a telepítő automatikusan beállítja a szükséges konfigurációt.

A választóképernyőn a fel/le nyilakkal és a sorszám megadásával is navigálhatsz. A telepítő a megerősítés előtt megmutatja a pontosan kiírandó MCP szekciót és a célfájlt. Az elfogadáshoz nyomd meg az Enter billentyűt. Ha adott ágenshez akarod közvetlenül telepíteni, add meg az asszisztens azonosítóját:

rocrate-mcp-server -i codex

Az elérhető azonosítók: codex, claude, opencode, kilo, roo, gemini és qwen. A létrehozott konfiguráció a standalone szervert a közös, felhasználónkénti RockIT socketen (~/.rockit/rocrate-mcp-server.sock) keresztül indítja.

A standalone szerver és a RocKIT alapértelmezés szerint ugyanazt a felhasználónkénti profil- és séma-tárolót használja: ~/.rockit. Másik közös gyökérhez állítsd be a ROCKIT_ROOT_PATH változót; a fájlnév- és szolgáltató- felülírásokhoz használd a ROCKIT_* profil-környezeti változókat.

A sikeres telepítést úgy ellenőrizheted, hogy elindítod az AI-asszisztenst, és listázod a beállított MCP-szervereket. Ehhez tipikusan az /mcp (például Codex és Claude Code esetében) vagy az /mcps (például OpenCode esetében) parancsot kell kiadni. A listában meg kell jelennie a rocrate nevű MCP-szervernek, amely a rocrate-mcp-server parancsot futtatja.

A rocrate-mcp-server használatához az AI-asszisztenst abban a könyvtárban indítsd el, ahol a RO-Crate-et létre szeretnéd hozni vagy szerkeszteni. Add meg az RO-Crate csomag létrehozásához a megfelelő utasítást, és az asszisztens automatikusan elkezdi használni a rocrate-mcp-server-t, követve az abban meghatározott munkafolyamatokat. Ha a RO-Crate szerkesztésekor az MCP-szerver nem indul el automatikusan, kérd meg kifejezetten az asszisztenst a következőre: „Használd az RO-Crate MCP-t!”.

A munkafolyamat végén az elkészült RO-Crate csomag (a ro-crate-metadata.json fájl) megnyitható az ARP AROMA szoftverben. Ezt általában automatikusan felajánlja az asszisztens. Ha nem, akkor csak kérd meg:

„   Nyisd meg az adatcsomagot az AROMA-ban”.

Ennek hatására az asszisztens ad egy URL-t, amelyre kattintva a böngészőben megnyílik az AROMA az éppen szerkesztett RO-Crate csomaggal, intelligensebb ügynök esetén pedig akár automatikusan meg is nyithatja a böngészőt.

Ha az AI-asszisztenst és a megnyitott AROMA böngészőablakot egymás mellé helyezed, akkor az asszisztenssel végzett módosítások azonnal megjelennek az AROMA-ban is, ahol ellenőrizhetők, illetve közvetlenül szerkeszthetők. Jó megoldás lehet egy asztali alkalmazásban, például a ChatGPT asztali alkalmazásában dolgozni, és az AROMA böngészőablakát az alkalmazáson belül megnyitni. A beépített böngésző olyan feladatok elvégzésére is használható, amelyek interaktív böngésző-hozzáférést igényelnek, például a „browser use”.

A szerver helyi vezérlőpultot is biztosít a MCP-tevékenység megfigyeléséhez a http://127.0.0.1:9393 címen, ahol különböző értékek is beállíthatók.

A rocrate-mcp-server használatával az adatcsomag az ARP Dataverse-be is feltölthető; ehhez csak kérd meg az asszisztenst. A feltöltéshez állítsd be a DATAVERSE_API_KEY környezeti változót. Ezt a https://repo.researchdata.hu/dataverseuser.xhtml?selectTab=apiTokenTab oldalon tudod beszerezni. Ha nincs ez a környezeti változó beállítva, az asszisztens is bekérheti.

A vezérlőpult Beállítások oldala lehetővé teszi a TAVILY_API_KEY, DATAVERSE_BASE_URL és DATAVERSE_API_KEY értékek megadását vagy felülírását az éppen futó MCP-folyamatban, valamint a sikeres Dataverse-feltöltések során keletkező RO-Crate ZIP-fájlok megtartásának beállítását. Alapértelmezés szerint a környezeti változók használatosak; a vezérlőpult beállításai a ~/.rockit/rocrate-mcp-settings.json fájlba kerülnek. Az API-kulcsok felülírásai a Beállítások oldalán törölhetők a környezeti változók visszaállításához. A fájl a beállított API-kulcsokat egyszerű szövegként tartalmazza, és ahol támogatott, csak a felhasználó számára engedélyezett hozzáféréssel rendelkezik.