npm Registry MCP Server

@yawlabs/npmjs-mcp

One click adds this to your local Yaw MCP config so it's available in every Yaw Terminal session. Or install manually below.

Run npm registry operations from Claude Code, Cursor, and any MCP client. 64 tools covering the full registry surface: package intelligence, security audits, dependency analysis, org/team management, and the write ops that normally fight you locally (npm deprecate, npm dist-tag, npm owner, npm unpublish).

Built and maintained by Yaw Labs.

Why this one?

Other npm MCP servers wrap npm search and call it done. This one doesn't.

• Full registry HTTP surface — 64 tools across reads, writes, orgs, teams, hooks, provenance, trusted publishers, and ops health. Not just npm view.
• Write ops that actually work in agents — npm_deprecate, npm_undeprecate, npm_dist_tag_set, npm_unpublish_version go directly to the HTTP API with a Granular Access Token that has 2FA bypass. No 2FA prompts, no --otp hunts, no ENEEDAUTH from a session-bound .npmrc. (Since 2026-07-31 npm requires an interactive 2FA challenge for owner, access, team membership and grant, org membership and token changes even with 2FA bypass; for those tools the error names the exact npm command a human runs.)
• Agent-aware failure surfacing — npm_check_auth and npm_publish_preflight detect a non-interactive context and hand back a human-runnable command, and every write error names what was sent and the npm CLI equivalent, instead of looping on unrecoverable errors.
• Safety by default — npm_unpublish_* requires confirm: true. npm_owner_remove blocks you from locking yourself out. npm_deprecate rejects a message over the registry's 1024-character limit before sending it.
• Ops playbook built in — npm_ops_playbook returns the canonical tool-vs-CLI-vs-CI decision matrix so your agent picks the right path on the first try.
• Tool annotations — every tool declares readOnlyHint, destructiveHint, idempotentHint, and openWorldHint, so MCP clients can skip confirmation on safe ops.
• No API key required for reads — search, packages, downloads, security, dep tree, licenses all work anonymously. Auth is opt-in via NPM_TOKEN.
• Instant startup — ships as a single bundled file with zero runtime dependencies. No 5-minute node_modules install.
• Input hardening — package names, scopes, versions, dist-tags, and team names are all regex-validated against npm's actual constraints. Defends against CRLF and path-traversal in URL construction.

Quick start

1. Create `.mcp.json` in your project root

macOS / Linux / WSL:

{

"mcpServers": { "npm": { "command": "npx", "args": ["-y", "@yawlabs/npmjs-mcp@latest"] } } }

Windows:

{

"mcpServers": { "npm": { "command": "cmd", "args": ["/c", "npx", "-y", "@yawlabs/npmjs-mcp@latest"] } } }

Why the extra step on Windows? Since Node 20, child_process.spawn cannot directly execute .cmd files (that's what npx is on Windows). Wrapping with cmd /c is the standard workaround.

2. Restart and approve

Restart Claude Code (or your MCP client) and approve the npm MCP server when prompted.

3. (Optional) Add your npm token for write operations

Read-only tools work without any setup. For write tools (deprecate, dist-tag, owner, team_*, org_member_*, unpublish, hook_*, access_set*, token_revoke), add NPM_TOKEN to the env block:

{

"mcpServers": { "npm": { "command": "npx", "args": ["-y", "@yawlabs/npmjs-mcp@latest"], "env": { "NPM_TOKEN": "npm_xxxxxxxxxxxx" } } } }

Use a Granular Access Token scoped to just the packages and orgs you want your agent to manage.

That's it. Now ask your AI assistant:

"Deprecate my-old-pkg 1.x with a pointer to v2"

"What's the dep tree for fastify look like three levels deep?"

"Audit express for known CVEs and tell me the fix"

"Who are the maintainers of next.js and when did each one last publish?"

Configuration

Runtime

The server ships a launcher that prefers the oam runtime and falls back to Node. The server itself is a pre-bundled ESM file using only node: builtins, so both paths behave identically — verified against the full MCP surface (handshake, all 64 tools, live registry calls) on each.

oam 0.19.1, the latest release, is the minimum. The launcher asks every oam binary it can find for its version and runs the newest one at or above it, never serves on an older oam, and falls back to Node when there is none (NPMJS_MCP_RUNTIME=oam turns that into a hard error). See Configuration for the details.

Falling back costs nothing: npm has already started Node to run the launcher, so the fallback is an in-process import() — no extra spawn, no extra startup.

oam is faster, but the launcher is not. Measured on windows-arm64, n=12 medians, spawn to first MCP initialize response:

npm bin entries are Node scripts, so reaching oam through one costs Node's startup *plus* oam's — more than oam saves. The launcher exists so npx users get oam automatically; it is not the fast path.

If you want oam's speed, point your MCP host straight at it and skip the launcher:

{ "mcpServers": { "npmjs": { "command": "oam", "args": ["run", "/abs/path/to/dist/index.js"] } } }

Benchmarking note: measure an installed oam (~/.oam/bin), never one out of a cargo target/ directory — a concurrent cargo build replaces the binary mid-run, and fresh bytes are cold where the node.exe you are comparing against is warm. oam is pre-alpha; re-measure on your own hardware.

oam's `--permission` sandbox is opt-in. Set NPMJS_MCP_SANDBOX=1 and the launcher spawns the server on oam (0.19.1 or newer) with:

• network granted to registry.npmjs.org:443, api.npmjs.org:443 and replicate.npmjs.com:443, plus the host and port of NPM_REGISTRY when it is set (its own port, or 443 / 80 for https / http). oam compares each request as host:port exactly, so every grant is scoped to the one port the server uses;
• environment granted to exactly the variables the server reads: NPM_TOKEN, NPM_REGISTRY, NPM_REQUEST_TIMEOUT_MS, NPM_RETRY_BACKOFF_MS and DEBUG;
• filesystem and subprocess access denied outright. The server reads no files at runtime and spawns nothing, so a dependency that suddenly wants either is stopped by the runtime instead of trusted — which matters for a process holding an NPM_TOKEN.

Under the default NPMJS_MCP_RUNTIME=auto, if no usable oam is found, or the one chosen fails to start, the server runs without the sandbox and says so on stderr. Set NPMJS_MCP_RUNTIME=oam as well to make that an error.

It is not the default because an incomplete grant fails silently rather than loudly. Measure