
This wraps the npm Registry API so Claude can search packages, fetch metadata, and pull version details without you leaving the conversation. No auth needed since it's hitting public npm endpoints. Handy when you're exploring dependencies, comparing package stats, or need to check what versions are available before updating a project. It's part of the Pipeworx gateway, which bundles 673+ data sources behind a single MCP connection. You can call tools directly or use the ask_pipeworx interface to query in plain English and let it route to the right endpoint. Runs over streamable HTTP, so setup is just dropping a URL into your MCP config.
npm MCP — wraps the npm Registry API (free, no auth)
Part of Pipeworx — an MCP gateway connecting AI agents to 1476+ live data sources.
| Tool | Description |
|---|---|
search_packages | Search npm for packages by keyword. Returns package names, descriptions, latest versions, publish dates, and publishers. Use when discovering libraries for a task. |
get_package | Get full details for a specific package: version, description, dependencies, homepage, repository, and license. Use after search_packages to evaluate a candidate package. |
get_downloads | Fetch npm download counts for a package over a named period (last-day, last-week, last-month, last-year) or a custom date range like "2024-01-01:2024-06-30". Returns total downloads, start date, and end date. |
list_versions | List published versions of an npm package with publish dates, dist-tags (latest/next/beta), and deprecation notices — answers "what versions of exist", "when was last published", "is the latest version deprecated". Returns versions newest-first, plus the package created/last-modified timestamps. |
get_version_info | Get metadata for a SPECIFIC version (or dist-tag) of an npm package: dependencies, dev/peer dependencies, engines, license, tarball + integrity hash + unpacked size, and any deprecation notice. Use for "dependencies of @", pinned-version audits, or inspecting a particular release. Version accepts an exact version (e.g. "4.18.2"), a dist-tag (e.g. "latest", "next"), or omit for the latest. |
Add to your MCP client (Claude Desktop, Cursor, Windsurf, etc.):
{
"mcpServers": {
"npm": {
"url": "https://gateway.pipeworx.io/npm/mcp"
}
}
}
tools/list at https://gateway.pipeworx.io/npm/mcp returns the tools in the table
above plus the shared Pipeworx meta-tools — ask_pipeworx,
discover_tools, search_within, remember/recall and the rest of the
gateway-wide set. So the tool count you see is larger than this table: a
single-pack endpoint currently lists roughly 30 shared tools alongside the
pack's own. The connection's initialize response states its exact scope, and
is the authoritative answer for a given day.
This is deliberate, not multiplexing by accident. The meta-tools are what let a
scoped connection answer a question this pack does not cover — via
ask_pipeworx, which routes across the whole catalog — without you adding a
second MCP server. There is currently no way to mount a pack endpoint without
them; if the extra schemas cost you more context than the routing is worth,
connect to the full gateway once rather than to several pack endpoints.
Or connect to the full Pipeworx gateway to get every pack's tools listed directly, instead of just this one's:
{
"mcpServers": {
"pipeworx": {
"url": "https://gateway.pipeworx.io/mcp"
}
}
}
Both URLs reach the same gateway and the same 1476+ data sources. The
only difference is which pack's tools are listed directly; ask_pipeworx
reaches all of them from either one.
Instead of calling tools directly, you can ask questions in plain English — this works on the pack endpoint above as well as on the full gateway:
ask_pipeworx({ question: "your question about Npm data" })
The gateway picks the right tool and fills the arguments automatically.
MIT