@pipeworx/new-jersey-code

New Jersey Statutes (N.J.S.A.) by citation — with amendment history where the source lists one — plus full-text topic search. Keyless.

Part of Pipeworx — an MCP gateway connecting AI agents to 1865+ live data sources. This is an independent, unofficial integration — not affiliated with, endorsed by, or published by the upstream provider.

Tools

• nj_statute_section(section) — full text of one N.J.S.A. section by citation (e.g. "2C:11-3", "39:4-50", "46:8-19"), with its amendment_history line parsed out when the section has been amended.

• nj_statute_search(query, limit?) — full-text search by topic/keyword ("security deposit interest", "landlord grounds for eviction"); returns ranked sections with citation + heading where the hit title parses as one.

Related MCP server: florida-code

Auth

Keyless.

Data sources

• https://lis.njleg.state.nj.us/nxt/gateway.dll — the New Jersey Legislature's statute-viewer subdomain (NXT4/Folio document publishing, IIS-hosted). This is a DIFFERENT legislature subdomain than the njleg.state.nj.us Next.js site most links point at — that one is bills/session focused and does not itself host statute text.

Shape of this source, because it is not a REST API

The viewer is session-cookie based, not a clean lookup-by-id endpoint:

1. f=xhitlist is a full-text search. It sets a session cookie and renders its hit list as plain server-side HTML (<a class="hit-doc-link" href="...">title</a> rows) — no JS execution needed to read it.

2. A citation lookup queries the EXACT quoted phrase "<citation> " (trailing space, inside the search). That trailing space is load-bearing: an unquoted citation free-text-matches thousands of sections that merely contain the same digits, and even the quoted phrase WITHOUT the trailing space still matches dotted/lettered children ("39:4-50.17", "39:4-50c") because "50" is a literal prefix of "50.17". The space forces the match onto a token boundary.

3. Each hit's href is an f=hitdoc$hitdoc_bm=... link. Resolving it (GET, with the search's session cookie, redirect: manual) 302s to the document's permanent, COOKIE-FREE path: /nxt/gateway.dll/statutes/<a>/<b>/<c>.

4. That path is fetched directly for the actual statutory text.

This is a session-cookie flow across two ordinary GETs — not the ASP.NET __VIEWSTATE postback pattern (there is no postback form anywhere in it), and it carries no CAPTCHA or other anti-bot wall. It has been reachable from a Cloudflare Worker at the rate one caller generates; it has not been load tested.

Amendment history is free; a prior point-in-time version is not

Every amended section's text ends with a trailing amended 1979, c.178, s.21; ...; 2017, c.150. (or, for a section's original enactment, L.<year>, c.<n>, s.<n>.) line. nj_statute_section parses that into amendment_history rather than leaving it buried in text — an original, never-amended section simply has none (null). This source does NOT expose a point-in-time/prior-version snapshot of a section; nj_statute_section says so explicitly in its historical_version field rather than silently omitting the capability.

Topic search quality

nj_statute_search ANDs every word in the query (this is the upstream engine's own behaviour, not something this pack adds) — a long sentence over-narrows; a few plain words ("landlord grounds for eviction") works better than a full question.

Stored copy (fleet #3771)

The statute viewer above is reachable but intermittent — measured over the 14 days before this shipped, 40 of 127 calls came back upstream_down (a direct connection to the origin timing out, not a block). So the statutes are ALSO stored, one object per Title, in the same shared statute store the other state-code packs use (New Jersey Legislature, New Jersey Statutes Annotated). A section found in the stored copy answers with no live request at all (served_from: "stored-copy", with data_as_of); a section not yet baked, or baked without text, falls back to the live flow above unchanged. Topic search gets the same treatment additively through a full-text index over the same stored copy — the live search is the fallback whenever the index has no match, so a caller never loses a capability this pack already had. The stored copy is built incrementally, Title by Title, not as one all-or-nothing full-code crawl — there is no bulk file or published table of contents for this source, so a Title's contents can only be learned by querying the search engine itself and reading back what it reports.

Quick Start

Add to your MCP client (Claude Desktop, Cursor, Windsurf, etc.):

{
  "mcpServers": {
    "new-jersey-code": {
      "url": "https://gateway.pipeworx.io/new-jersey-code/mcp"
    }
  }
}

What this endpoint actually serves

tools/list at https://gateway.pipeworx.io/new-jersey-code/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 1865+ data sources. The only difference is which pack's tools are listed directly; ask_pipeworx reaches all of them from either one.

No MCP client? Call it over HTTP

curl -X POST https://gateway.pipeworx.io/v1/tools/nj_statute_section \
  -H 'Content-Type: application/json' \
  -d '{"section":"39:4-50"}'

No account needed for the first calls. Inspect any tool: GET https://gateway.pipeworx.io/v1/tools/nj_statute_section. Find one: POST https://gateway.pipeworx.io/v1/tools/search_packs with {"query":"..."}.

Run it from npm

@pipeworx/mcp-new-jersey-code is a local stdio bridge to the Pipeworx gateway: your MCP client starts it with npx, and it forwards every call to the same https://gateway.pipeworx.io/new-jersey-code/mcp endpoint shown above. It needs network access; there is no offline mode.

{
  "mcpServers": {
    "new-jersey-code": {
      "command": "npx",
      "args": ["-y", "@pipeworx/mcp-new-jersey-code"]
    }
  }
}

Without a key it calls anonymously, capped at 50 calls a day. A free account raises that to 200 a day: get a key at https://pipeworx.io/signup and add "env": { "PIPEWORX_API_KEY": "<your key>" } to the server entry. The key is sent as a Bearer token, the same way a hosted connection sends it, so calls are metered and billed exactly as they are there.

It lists only this pack's tools by default. Set PIPEWORX_META_TOOLS=1 to also list ask_pipeworx and the other gateway-wide tools.

Behind an HTTPS proxy: Node's built-in fetch does not read HTTPS_PROXY by default. Newer Node releases honour it when NODE_USE_ENV_PROXY=1 is set; check your Node version's documentation.

Using with ask_pipeworx

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 New Jersey Code data" })

The gateway picks the right tool and fills the arguments automatically.

More

• Docs and guides

• pipeworx.io

License

MIT