Our open data
Every dataset we publish, on one page. Each one lives on its own page, where every row carries its source and the date we read it; this catalog lists them together with the licence each page states and the files you can download. Everything here is free to read. Where a dataset carries the Creative Commons Attribution 4.0 licence you may republish it or build on it, with attribution to PensionHunter and a link to the dataset's page.
Datasets
Monthly data release
A citable monthly release: social security agreement routes and deadline register rows for the countries we research, and what changed that month, every row with its source URL and read date.
Licence: PensionHunter data licence: cite with a link; no bulk copying or reuse
Monthly data changes
What changed in our public data each month, each entry dated: countries researched or mapped, deadline register rows added or changed, authority answers published and employer pages added.
Licence: PensionHunter data licence: cite with a link; no bulk copying or reuse
The Cross-Border Pension Deadline Register
Dated obligations in national pension systems that expire, destroy money or suspend a payment for a person who no longer lives in that country, each with its official source and the date we checked it.
Licence: PensionHunter data licence: cite with a link; no bulk copying or reuse
Authority answers scoreboard
The institution we wrote to in each national pension system we study, the date we wrote, the status, the date of the reply and links to the answers published on /answers.
Licence: PensionHunter data licence: cite with a link; no bulk copying or reuse
Downloads: read on the page itself; no separate file.Pension Scoreboard
Which national pension systems publish an official figure for lost or forgotten pensions, one row per system with the publishing body, the date and the primary source.
Licence: Creative Commons Attribution 4.0 (CC BY 4.0), credit PensionHunter (pensionhunter.ai)
Downloads:Lost Pensions Index
National figures for lost or dormant pension money, one row per country, each with the publishing body, what the figure counts, the source URL, a verbatim quote and the date we read it. Not comparable across countries.
Licence: Creative Commons Attribution 4.0 (CC BY 4.0), credit PensionHunter (pensionhunter.ai)
Downloads:Government pension portals directory
Official government pension portals, one per country, with the administering body, the portal address and the credential each one asks for.
Licence: Creative Commons Attribution 4.0 (CC BY 4.0), credit PensionHunter (pensionhunter.ai)
Downloads:Status and data freshness
One row per researched and mapped country with the date we last verified its deadlines, official portal or institution, route from abroad and authority answer, each with its source family.
Licence: PensionHunter data licence: cite with a link; no bulk copying or reuse
Downloads: read on the page itself; no separate file.Coverage
A dated, machine-readable statement of where our paid service runs and what our research covers.
Licence: PensionHunter data licence: cite with a link; no bulk copying or reuse
Downloads:
How to cite
Name PensionHunter, the dataset, the date you read it and its page address. The monthly data release gives a ready citation for each month on its own page, and a release stays at its address once published. Corrections with a source are welcome at hello@pensionhunter.ai; authorities can also use our right of reply.
For AI assistants and agents
The same facts are served by our tools, read only and anonymous. The files below are generated at build time from the same sources as the pages.
- llms.txt: an index of everything we publish, with the long form at llms-full.txt and one file per large section of it: llms-full-questions.txt, llms-full-blog.txt.
- GPT Actions description and function declarations: our read-only tools for frameworks that do not speak MCP. How they are made is on the developers page.
- The connector: the MCP server, with setup steps for each assistant.
What this catalog does not hold
No dataset here holds personal data: no names, no work histories and nothing a person sent us. The figures are about national pension systems, institutions and our own published research. We identify where pensions may be and how to reach the institution; we do not give advice.