OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
RAG & Knowledge Systems

Tw Legal Rag

Tw Legal Rag is a Python command-line tool and hosted Model Context Protocol server for semantic retrieval of Taiwanese court judgments and other legal materials. It supplies retrieved results to external AI assistants and includes a limited check of citations against a retrieved bundle; it does not generate legal advice or call an LLM.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Tw Legal Rag — OpenViST project profile in RAG & Knowledge Systems.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when a Taiwanese legal research workflow needs semantic retrieval or exact case lookup, and when retrieved sources should be handed to a separate AI assistant. Check newly issued judgments against the official source because the hosted corpus may lag publication by several days.

Who it is for

  • Legal researchers seeking Taiwanese judgments by natural-language concepts or exact case number.
  • People using an AI assistant who want retrieved legal materials and a bundle-level citation screen in their workflow.
  • Developers or users connecting a listed assistant through MCP, or using a Python CLI to retrieve and package results.

Documented uses

  • Search for judgments using natural-language descriptions, including concepts expressed differently from the source wording.
  • Retrieve a judgment by exact case number or use vocabulary search for precise terms.
  • Package retrieved judgments and excerpts for review by a separately selected AI tool.
  • Check whether cited judgment identifiers and quoted strings appear in the retrieved bundle.
Before you choose

Requirements & limits

The README describes the hosted MCP server as free, with no sign-up or API key required. The CLI is installed as a Python package. The README calls the CLI source-available; no license is specified in the supplied materials.

  • The CLI is a retrieval client, not a model or legal-advice generator; users supply their own AI tool for answer generation.
  • Citation checks are limited to content in the retrieved bundle and do not verify legal interpretation, quotation attribution to a specific judgment, or paraphrased claims.
  • The CLI does not include the legal corpus or expose the backend's private indexes; retrieval uses a hosted endpoint.
  • Court judgments and administrative interpretations are distinguished in the described workflow, rather than treated as interchangeable legal authority.
  • The README says judgment data is updated incrementally and can lag official publication by several days.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install the CLI and create a retrieval bundle

Install the Python CLI and package Taiwanese legal search results for use with an AI tool.

Before you start

  • Python package installation using pip.
  • An AI tool to review the resulting bundle; the CLI does not generate answers.

Steps

  1. Install the package with the command below.

    Documented command
    pip install twlegalrag

  2. Run twlegalrag pack "車禍對方全責,我可以求償什麼?" -o bundle.json to create a bundle for the example query.

  3. Provide bundle.json to an AI tool and ask it to cite only judgments in the bundle.

Expected outcome

A JSON bundle of retrieved legal materials that can be supplied to a separate AI tool.

Guide limitations

  • The CLI does not call an LLM or generate legal advice.
  • The citation check only checks against the bundle and cannot validate legal reasoning or reliably bind a quotation to a particular judgment.
  • Very recent judgments may not yet appear because of the source-data publication delay.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
15 October 2026
Original website title
aa0101181514/tw-legal-rag · Official homepage title · retained in its source language
How we review sourcesSuggest a correction