What is this?

HUDOC Researcher is a paragraph-level companion to the European Court of Human Rights. It indexes every paragraph of 20,060 judgments (up to 8 October 2026), each with the Court’s own paragraph number and the part of the judgment it belongs to, so that a search lands on the passage you will cite — and then sends you to HUDOC, the authoritative source, for the full text.

Search — full text with HUDOC’s operators, the parts of the judgment to search in, filters, and each judgment’s Cites / Cited by.
Semantic Search (experimental) — describe a case in plain language.
Check — paste a text: every citation, paragraph number and quotation is checked against the judgments.
Workspace — paragraphs and searches you save, with your notes, kept in your browser.
Statistics — the corpus in charts.
AI assistant (MCP) — the same search from Claude or another assistant, quoting paragraphs verbatim.

How the data was built and checked is set out in the Methodology.

Authors

Łukasz Szoszkiewicz

Adam Mickiewicz University, Poznań

Sebastian Marcisz

Adam Mickiewicz University, Poznań

How to cite

Szoszkiewicz, Ł., & Marcisz, S. (2026). HUDOC Researcher — ECtHR case-law search and RAG [Computer software]. Zenodo. https://doi.org/10.5281/zenodo.21319703

DOI 10.5281/zenodo.21319703

The DOI is Zenodo’s record for all versions of the software; each release is archived there. When you rely on a passage, cite the judgment itself as well — its HUDOC record and paragraph number.

Data, licences and code

Source: the judgments and their metadata come from HUDOC, the Court’s database; the one-line summaries of Grand Chamber cases from the Registry’s Table of all Grand Chamber judgments and decisions, which the Registry states does not bind the Court.
Software is released under the PolyForm Noncommercial License 1.0.0 — research, education, non-profit and personal use are permitted; commercial use needs a separate licence. The code of this website is at github.com/lszoszk/ECHR-Dashboard, and every release is archived on Zenodo; the data pipeline and the search API are not public and are available for research on request.
The curated dataset (the segmentation into paragraphs, paragraph numbers, section labels, normalised metadata and citation links) is licensed under CC BY-NC-SA 4.0 — academic and non-commercial reuse is welcome with attribution; commercial use requires permission. It is published on Hugging Face.
The judgments themselves are © Council of Europe / European Court of Human Rights and remain under the Court’s terms for HUDOC content; this project does not re-license them.

Bug reports and suggestions are welcome. For commercial licensing, write to l.szoszkiewicz@amu.edu.pl.

Use it from your AI assistant (MCP)

The corpus is also available as a Model Context Protocol server, so an AI assistant (Claude, or any client that supports MCP) can search the judgments and quote paragraphs verbatim, each with its case name, application number, date and paragraph number, instead of paraphrasing from memory. It offers seven tools: echr_search (paragraph search with the same filters as Search), echr_get_case, echr_get_paragraphs, echr_cited_by and echr_cites (the citation links), echr_facets and echr_stats.

1. Add the connector (nothing to install)

The server is hosted with this site. Its address:

https://150.254.115.204/echr-api/mcp

Claude (claude.ai, Desktop, mobile): Settings → Connectors → Add custom connector, paste the address, no sign-in needed. On Team and Enterprise plans an owner adds it for the organisation.

Claude Code:

claude mcp add --transport http echr https://150.254.115.204/echr-api/mcp -s user

Any other assistant that accepts remote MCP servers (Streamable HTTP): the same address.

2. Add the research skill (optional, for Claude)

A skill that teaches Claude how to use the tools well: find the case-law on an issue, identify the leading case, quote the Court verbatim with OSCOLA pinpoints (Kudła v Poland [GC] App no 30210/96 (ECtHR, 26 October 2000), para 152), tell the Court’s reasoning from separate opinions and facts, and check the ECtHR citations in a draft.

Claude: download echr-research.zip, then Settings → Capabilities → Skills → Upload skill.
Claude Code: copy the skills/echr-research folder into ~/.claude/skills/.

3. Ask

For example: “Find Grand Chamber paragraphs on the margin of appreciation under Article 10 and cite them with paragraph numbers” or “Check the ECtHR citations in this paragraph.”

Run it yourself

The same server can run on your computer (Node.js 18 or newer), for instance if your assistant only takes local servers:

Download it and install its dependencies:

git clone https://github.com/lszoszk/mcp-echr.git
cd mcp-echr
npm install

Claude Desktop: add this to claude_desktop_config.json (Settings → Developer → Edit Config), with the folder you cloned into, then restart Claude:

{
  "mcpServers": {
    "echr": {
      "command": "node",
      "args": ["/ABSOLUTE/PATH/TO/mcp-echr/src/index.js"]
    }
  }
}