russellbrenner/jurisd — Jurisd (formerly auslaw-mcp) is a local-first MCP server and CLI giving any AI agent grounded Austra…

Description and use

**Brief Summary**
Jurisd (formerly auslaw‑mcp) is a TypeScript‑based, local‑first MCP server and CLI that gives AI agents authoritative access to Australian and New Zealand legal sources. It bundles AustLII and jade.io retrieval, deterministic provision lookup, an offline citation graph, fast AGLC4 linting, and optional semantic‑search/graph‑recall over user‑supplied corpora, all behind a standard Model Context Protocol (MCP) interface.

**Value**

• **Grounded legal knowledge** – AI assistants can query up‑to‑date statutes, case law, and regulations rather than hallucinating, which is essential for compliance‑heavy domains.
• **Standardised integration** – By exposing a clean MCP/CLI/SDK surface, Jurisd lets developers plug any MCP‑compatible model (e.g., Claude, GPT‑4o) into a reliable legal‑research back‑end with minimal custom code.
• **Offline resilience** – All core data (citation graph, AGLC4 rules) lives locally, so the service continues to work even when external sites are down or rate‑limited.
• **Extensible RAG** – The optional semantic‑search layer lets teams augment the built‑in legal corpus with proprietary documents, creating a unified “legal‑plus‑internal” knowledge base.

**Practical Adoption Path**

1. **Spin up the server** – Clone the repo, run npm install && npm start (or use the provided Docker image) to launch a local MCP endpoint.
2. **Configure data sources** – Add AustLII/jade.io API keys (if needed) and point the server at any additional document collections you want to index for semantic search.
3. **Connect your model** – In your AI‑agent code, use the MCP SDK or simple HTTP calls to send search, lookup, or lint requests to the Jurisd endpoint.
4. **Iterate & test** – Verify responses against known legal citations; adjust fallback settings or indexing parameters.
5. **Deploy** – Package the server as a container or a managed VM; because it’s local‑first, scaling is just a matter of adding more instances behind a load balancer.

**Production Readiness**

• **Activity & community** – 24 ★, 8 forks, recent commits (as of 2026‑06‑28) and a clear TypeScript codebase indicate an active project.
• **Maturity** – Core legal‑research features (AustLII, jade.io, citation graph, AGLC4 linting) are stable; the optional semantic layer is modular and can be disabled if not needed.
• **Integration simplicity** – Exposes a well‑documented MCP API, CLI, and SDK, making it straightforward to evaluate and embed in existing pipelines.
• **Risk considerations** – No major metadata or licensing red flags yet, but a final security audit and confirmation of maintainer commitment are advisable before a large‑scale rollout.

Overall, Jurisd is a high‑readiness OSS component for any organization that wants to ground AI assistants in authoritative Australian/New Zealand law while retaining the flexibility to add private corpora.

Problem

russellbrenner/jurisd helps connect AI assistants to real tools and data through a standard protocol.

Outcome

High for an OSS candidate: recent activity, adoption, and ecosystem signals are strong enough for a serious pilot.

Workflow

• connect AI agents to tools
• ship Model Context Protocol servers
• standardize integrations

Evidence

jurisd (formerly auslaw-mcp) is a local-first MCP server and CLI giving any AI agent grounded Australian and New Zealand legal research: AustLII and jade.io retrieval with resilient fallback, deterministic provision lookup, offline citation graph, fast AGLC4 linting, plus optional semantic search and graph recall over your own local corpora.

Source