About Luni
Useful web data should come with a source and a clear account of the work. That is the starting point for Luni.
A focused web-data API
Luni gives developers a small set of operations for reading the public web: scrape a page, map URLs, search for sources and crawl a bounded part of a website. Company research is available only to invited enterprise customers. The workspace brings keys, requests, jobs and credit usage into one place.
The API is intended as a building block. Your application decides what to collect, where to store it and how to use it. Luni supplies extracted content and operation results rather than an entire research, retrieval or sales platform.
Evidence travels with the content
A document is more useful when it can be traced to its source. Keep source metadata with the text you ingest, and keep observations separate from conclusions produced later by a person or model.
The same principle applies to company research: a published contact and its source are an observation, not an independent verification of a person or inbox. Missing fields should stay explicit rather than being filled by assumption.
Bounded work and visible costs
Set a page limit before a crawl, inspect actual request history and distinguish a retry from a new observation. Luni reserves a maximum budget and settles completed work, releasing unused reservations.
The current product is in beta. Public pages describe the supported contract and availability conditions; they do not promise universal site coverage, a production uptime guarantee or features outside that contract. Start with a representative sample of your own sources.