Gadoes builds a scraper framework for jurisdictions the APIs forgot
When there's no off-the-shelf data feed for a body of law, scrape it yourself - and make that repeatable.
Most legal-AI tools lean on tidy commercial APIs to pull in case law. That works for the big English-language jurisdictions and falls apart everywhere else. Gadoes has built a small framework inside the dispumike fork that treats web-scraping as a first-class source: a base scraper contract, a freshness tracker that knows when corpora go stale, and an adapter that makes the scraped output look identical to a proper data feed downstream.
To prove the pattern works, the fork ships two concrete implementations: italaw, covering Italian law, and ICSID, the international investment-arbitration awards database. Neither has a friendly API. Both now plug into the rest of the system as if they did. The framework is deliberately separable from the scrapers themselves, so other forks can adopt the plumbing without inheriting Italy or ICSID.
Spotted something wrong? Or know the PR text has fresher detail than the writeup above?