@pipeworx/alaska-code

Alaska Statutes (Alaska Code) — current statutory text by citation, and topic/keyword search over official chapter catchlines, sourced live from www.akleg.gov.

Part of Pipeworx — an MCP gateway connecting AI agents to 1861+ live data sources. This is an independent, unofficial integration — not affiliated with, endorsed by, or published by the upstream provider.

Tools

• ak_statute({citation}) — full text of one section by citation (e.g. "4.11.010" / "AS 04.11.010", alcohol-license requirements; "11.41.100", murder), with the official catchline and title/chapter context.

• ak_search({query, limit?}) — topic/keyword search over the 942 live chapter catchlines across all 47 titles, e.g. "alcohol license", "murder". Not full-text search, and not section-level — akleg.gov exposes neither keylessly (see "Data sources").

Related MCP server: florida-code

Auth

Keyless. No signup, no activation gate.

Data sources

https://www.akleg.gov/basis/statutes.asp — the Alaska State Legislature's own public statute viewer. It is driven as a GET AJAX endpoint (the page's own scripts/statutes.js calls it this way):

• ?media=print&secStart={citation}&secEnd={citation} — statute text for the half-open citation range [secStart, secEnd). Citation shape is title.chapter.section, e.g. 4.11.010. ak_statute always requests a single section (secEnd = the next section number) and gets back exactly one statute div block with no preceding Article/Chapter heading markup.

• ?media=js&type=TOC&title={N} — the chapter list (number + official catchline) for Title N. Baked once into src/ak-index-data.ts by scripts/bake-index.mjs (47 requests, one per title).

Verified live from both a laptop and a wrangler dev --remote Worker on the prod account on 2026-10-07 — identical byte lengths from both origins. Keyless, access-control-allow-origin: *.

The "not found" signal

A secStart/secEnd range that matches nothing — a nonexistent title, a nonexistent chapter, or a genuine gap number inside a real chapter — always comes back HTTP 200 with a zero-length body and no FirstSec/LastSec response headers (all three cases verified live). A real section returns a non-empty body. ak_statute branches on body length, not status code — a non-200 status, or a non-empty body that does not parse as a statute fragment, is thrown loud (httpError / a parse-error Error), never folded into found: false.

Not the same wall as Arkansas or Mississippi

Both of those states' own legislature sites link their code out to lexisnexis.com/hottopics/<state>code/, which redirects to an advance.lexis.com/container?config=... SPA shell requiring a Lexis session — confirmed live and at the edge (same destination host, same 3,696-byte shell, for both states). Alaska's own site serves the statute text itself, server-side, with no redirect to any third-party platform.

Two capabilities available, two not

• Citation lookup — ak_statute.

• Topic/keyword search — ak_search, over official CHAPTER catchlines only (942 of them). akleg.gov's type=TOC call returns "No Sections" for a dotted (title.chapter) argument rather than a per-section listing (verified live) — unlike Idaho, there is no section-level catchline feed to bake.

• Amendment history — NOT AVAILABLE. Unlike Idaho, the statute fragment this endpoint returns carries no trailing history citation (checked against AS 11.41.100, a frequently amended section with multiple [Repealed, ...] sub-paragraphs and nothing after them).

• Historical version (text as it read before an amendment) — NOT AVAILABLE, for the same reason as idaho-code and north-dakota-code: Session Laws are organized by session chapter number, not by Code citation.

Encoding

akleg.gov declares charset=ISO-8859-1 and means it literally — the one non-ASCII byte observed in testing (0xA7, the section sign § in a "[Repealed, § 44 ch 102 SLA 1980.]" note) decodes correctly as Latin-1. src/index.ts decodes the raw response bytes with TextDecoder('iso-8859-1'), not UTF-8.

Baked index refresh

src/ak-index-data.ts is GENERATED by scripts/bake-index.mjs from the 47 title TOC pages (942 chapters captured, ~70KB raw — small enough to stay a plain module import rather than going through the pack-baked-indexes.json / KV path south-dakota-code and idaho-code use for multi-MB tables). Re-run it after a legislative session's codification pass:

node scripts/bake-index.mjs > src/ak-index-data.ts

Quick Start

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

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

What this endpoint actually serves

tools/list at https://gateway.pipeworx.io/alaska-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 1861+ 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/ak_statute \
  -H 'Content-Type: application/json' \
  -d '{"citation":"4.11.010"}'

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

Run it from npm

@pipeworx/mcp-alaska-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/alaska-code/mcp endpoint shown above. It needs network access; there is no offline mode.

{
  "mcpServers": {
    "alaska-code": {
      "command": "npx",
      "args": ["-y", "@pipeworx/mcp-alaska-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 Alaska Code data" })

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

More

• Docs and guides

• pipeworx.io

License

MIT