About
Making public data legible to assistants
A lot of the most useful public data — safety reports, recalls, filings — lives behind clunky search forms and bulk downloads. PolarisRay wraps those datasets in MCP servers so an AI assistant can query them conversationally and cite the source. It's built for healthcare and life-sciences teams — clinical researchers, pharmacovigilance and device-safety groups, and regulatory and competitive-intelligence analysts.
Principles
- Source-faithful. Hits carry an official
urlback to the primary record when one exists. We don't editorialize the data. - Open by default. Public data stays free to query. No account required for public sources.
- Transparent limits. Each dataset's caveats and update cadence are documented on its page.
Project status
PolarisRay is free and in an early beta release. Currently, it isn't a commercial product — there are no ads, no paywalls, no accounts, and no plans to sell your data. The website uses no cookies, tracking, or analytics. Because it's still in beta, datasets, tools, and endpoints may change or be temporarily unavailable while things are refined. If you rely on a source, please double-check against the official dataset.
Get in touch
Want a dataset added, or found something off? Reach out at hello@polarisray.com.