UHRI+
Search and analytics across recommendations from UN human rights mechanisms
UHRI+ turns the Universal Human Rights Index into an exploratory research environment for comparing recommendations addressed to States by the Universal Periodic Review, Treaty Bodies, and Special Procedures.
What it does
- Searches recommendations and observations by country, mechanism, theme, affected persons, SDG, year, and annotation type.
- Supports comparative analysis across States and mechanisms rather than one-document-at-a-time browsing.
- Includes a labeling workflow and community taxonomies for developing reproducible research classifications.
- Keeps every result tied to its UN document symbol and original metadata.
UHRI+ recommendations are available through MCP
The same mcp-unhrdb connector exposes dedicated tools for recommendation search, record lookup, and facet discovery. The recommendations corpus remains separate from UNHRDB paragraphs, preventing different source types from being blended.
Why it matters
The UHRI contains a uniquely rich record of how UN mechanisms address human rights problems across countries and time. UHRI+ makes this material suitable for systematic legal and social-science research while preserving the wording and provenance of each recommendation.
Data access
The cleaned and normalised corpus of 267,671 recommendations is available in Parquet format on Hugging Face. Each record retains the UN document symbol, mechanism, country, year, themes, affected persons, SDGs, and original recommendation text.
The dashboard software has a stable Zenodo concept DOI. For reproducible software citation, cite the DOI of the exact release used; for analysis of the recommendations themselves, cite the dataset and record its version, query, and filters.