@pipeworx/at-ris

Austria's consolidated federal law and case law — full text, by provision or by decision, with in-force dates kept separate from publication dates: search and read a §/Article/Annex of a federal statute as of any date, and search and read a decision of the constitutional court (VfGH), the supreme administrative court (VwGH) or the ordinary courts (OGH/OLG/LG/BG), by Geschäftszahl or ECLI.

Part of Pipeworx — an MCP gateway connecting AI agents to 1863+ live data sources. This is an independent, unofficial integration — not affiliated with, endorsed by, or published by the upstream provider.

Tools

• at_search_legislation(query?, titel?, gesetzesnummer?, in_force_on?, limit?, offset?) — search consolidated federal law (Bundesrecht konsolidiert) by full-text words, law title, or exact law number. Returns matching provisions with their law, citation, in-force dates, subject index and the official RIS page.

• at_get_provision(law, paragraph, provision_type?, as_of?, text_offset?) — the original-language (German) text of one provision, as of a given date. law accepts a Gesetzesnummer or a title/abbreviation (resolved first). Long text is paged via text_offset/text_has_more/text_next_offset.

• at_get_legislation_document(document_id, text_offset?) — fetch one Bundesrecht konsolidiert document's text directly by its RIS id (from the two tools above), without re-resolving the citation.

• at_search_case_law(court, query?, geschaeftszahl?, date_from?, date_to?, doc_type?, limit?, offset?) — search case law in exactly one court family (vfgh, vwgh or justiz) by full-text words, case number and/or date range. Defaults to decision texts; doc_type: "rechtssatz" searches headnote summaries instead.

• at_get_decision(document_id?, ecli?, geschaeftszahl?, court?, text_offset?) — the original-language full text of one decision, by document id, ECLI (court inferred automatically) or Geschäftszahl (court required). Returns decision date, legal/subject area, headnote and norms cited alongside the text, paged the same way as at_get_provision.

Related MCP server: German Law MCP

Auth

Keyless.

Data sources

• Bundeskanzleramt (Austria), Rechtsinformationssystem (RIS) — Bundesrecht konsolidiert (consolidated federal law): every federal statute and regulation kept up to date with later amendments folded in, broken into provisions (§/Article/Annex), each with its own historical versions ("Fassungen") and in-force dates. These consolidated texts are explicitly informational, not the legally binding form (that is the separately authenticated Bundesgesetzblatt, not covered by this pack).

• Bundeskanzleramt (Austria), Rechtsinformationssystem (RIS) — Judikatur (case law): decisions of the Verfassungsgerichtshof (VfGH, constitutional court, from 1980), the Verwaltungsgerichtshof (VwGH, supreme administrative court, from 1990) and the ordinary courts (OGH plus OLG/LG/BG, "Justiz").

• A record's own RIS page is linked back from search/get results (official_url / document pages) for citation — that is the record, not a route.

Notes

• Keyless, live proxy, low-volume by design. RIS asks to be notified before high-volume harvesting. This pack never bulk-crawls: every tool makes exactly one upstream search request, and the two "get" tools make at most one further request to fetch that one document's text. Nothing is baked into the pack or mirrored.

• Search never returns body text. RIS's search API returns metadata and links to a document's Xml/Html/Rtf/Pdf, never inline text or a snippet. Getting the actual original-language text is always a second fetch, done inside at_get_provision / at_get_decision / at_get_legislation_document from the URL the search gave — never a URL constructed from guesswork, except in the two "by id" tools, which use the one confirmed-stable path shape for a caller who already holds an id from an earlier search.

• The upstream JSON is XML-derived: any field that can repeat comes back as {"item": "x"} for one value or {"item": ["x","y"]} for several — never a bare array, and the key is absent (not []) when there are none.

• Fassung (version) resolution is always explicit. A provision can have several historical versions, each with its own in-force window. This pack always sends an as-of date (defaulting to today) to resolve to exactly the one version valid then, rather than returning every version and asking the caller to pick, so in_force_from/in_force_until/publication dates on a response are for the one version actually returned.

• Case law id prefixes carry the court, which is how at_get_decision infers it from document_id or ecli without an extra round trip: VfGH decisions start JFT, VwGH JWT, and Justiz (OGH/OLG/LG/BG) JJT (full text) or JJR (headnote/Rechtssatz). A bare geschaeftszahl does not reliably carry this, so that path requires court explicitly.

• "Justiz" is the ordinary-courts family, not only the OGH — it also carries OLG (higher regional courts), LG/BG (regional/district courts) and a few specialised senates. at_search_case_law(court: "justiz", ...) searches all of them together; RIS does not let a caller narrow to "OGH only" within that application.

• Long provisions and decisions are paged by character offset (text_offset/text_has_more/text_next_offset), never silently truncated — some Justiz decisions run well past a single response's worth of text.

Quick Start

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

{
  "mcpServers": {
    "at-ris": {
      "url": "https://gateway.pipeworx.io/at-ris/mcp"
    }
  }
}

What this endpoint actually serves

tools/list at https://gateway.pipeworx.io/at-ris/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 1863+ 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/at_search_legislation \
  -H 'Content-Type: application/json' \
  -d '{"query":"Datenschutz","limit":5}'

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

Run it from npm

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

{
  "mcpServers": {
    "at-ris": {
      "command": "npx",
      "args": ["-y", "@pipeworx/mcp-at-ris"]
    }
  }
}

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 At Ris data" })

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

More

• Docs and guides

• pipeworx.io

License

MIT