Where the cases come from

• Supreme Court of Canada — published at scc-csc.ca and decisions.scc-csc.ca. Decisions are published openly by the Court.
• Federal Court / Federal Court of Appeal — decisions.fct-cf.gc.ca. Decisions are published openly by the Courts.
• Tax Court of Canada — decision.tcc-cci.gc.ca.
• Provincial appellate courts (roadmap, not yet ingested) — ontariocourts.ca, bccourts.ca, albertacourts.ca, citoyens.soquij.qc.ca, and equivalents. Phase 1 is a federal corpus; provincial appellate decisions are the next coverage expansion.

Most of the 28,000+ imported cases in caselaw · Canada came in via the Refugee Law Lab's open-access bulk dataset, hosted on Hugging Face: refugee-law-lab/canadian-legal-data.

The Refugee Law Lab is a Pro Bono Students Canada / York University research initiative that mirrors openly published Canadian legal materials and republishes them as bulk-downloadable datasets to enable access-to-justice research.

The dataset includes Supreme Court of Canada (1877–present), Federal Court of Appeal (2001–present), Federal Court (2001–present), Tax Court of Canada (2003–present), Canadian Human Rights Tribunal (2013–present), Refugee Protection Division (RPD), Refugee Appeal Division (RAD), Refugee Law Lab Reporter (RLLR), Social Security Tribunal (SST), and federal legislation/regulations.

We import all decision subsets. The IRB tribunal subsets (RPD, RAD, RLLR) and SST appeals are already redacted at source: the IRB and SST publish decisions with claimant identifiers anonymised under the Immigration and Refugee Protection Regulations and Department of Employment and Social Development Act. We mirror those redacted versions verbatim. If you spot any decision where redaction has slipped, email us and we will remove it within 24h.

Licence: CC BY-NC 4.0 (Creative Commons Attribution-NonCommercial 4.0 International). Attribution: Refugee Law Lab, refugeelab.ca/bulk-data. Use of caselaw · Canada is free during beta; if a paid tier is introduced for premium features, the NC clause applies and we will at that point either license bulk redistribution from the upstream source or remove the relevant subsets.

CanLII. Excellent public-interest database. CanLII's terms of use forbid bulk download and redistribution. We have not scraped CanLII at any point. Any future commercial relationship with CanLII will follow their licensing terms.

Briefs are AI-generated study aids, clearly labelled as such on every page and anchored to the scraped source text. They are not official summaries and have not been reviewed by a lawyer. Citations are validated against scraped metadata; failed-validation briefs are hidden from the library until a human approves them.

For all other cases, the case page shows the court-reporter headnote (where one exists) or the first portion of the judgment text. The judgment text and source URL are always linked at the bottom of each page so readers can verify against the authoritative original.

Briefs and notes here are study aids. They are not legal advice. For legal advice, talk to a member of your provincial law society.