Skip to content

Nano V2.1 is built.

Model family
Request access

Aarvi AI · Coverage

4,118,402 sites, and a number that changes every hour.

Collection is the part nobody demos and everything depends on. This page describes what is in the corpus, how fresh each channel is, and what the corpus does not claim to know.

4.1M+
sites indexed
24/7
collection window
3
query channels
2
cloud regions

Corpora

Six sources of evidence.

Each corpus keeps its own collection schedule, its own reliability profile and its own caveats. They are queried together and cited separately.

  • 01

    Dark web sites

    4.1M+ indexed

    Forums, marketplaces, paste dumps, leak pages and directories. Each capture is dated so freshness is never implied.

  • 02

    Threat actors

    aliases · campaigns

    Actor records with aliases, associated infrastructure, posted material and the campaigns they appear in.

  • 03

    Indicators

    domains · IPs · hashes

    Extracted from collected content and verified feeds, with first-seen, last-seen and the source that asserted it.

  • 04

    Telegram

    configured sources

    Monitored channels and groups, collected and indexed so a query never depends on a live scrape.

  • 05

    Leak sites

    ransomware victim pages

    Victim entries, claims and deadlines, linked to the actor that published them.

  • 06

    Open sources

    news · advisories · records

    Vulnerability news, public advisories and registries used to corroborate what the closed sources claim.

Freshness

How old is the answer you just got?

Three channels, three different meanings for the word “current”. The interface states which one produced a result rather than flattening them together.

Freshness characteristics of each coverage channel
ChannelWhat it holdsFreshness
Live dark webA fresh crawl of configured, reachable sources for the phrase you gave itMinutes to hours, source-dependent
Indexed dark webPages captured by earlier crawls and served by dwsearch.onlineFollows crawl and index lag
TelegramMessages already collected from configured channels and groupsIn development
Intelligence storeActors, campaigns and indicators maintained by scheduled jobsDaily to weekly, by source
Open sourcesAdvisories, vulnerability news and public records used for corroborationPer source schedule

Limits

What we do not claim.

  • Search runs over configured and reachable sources. It is not an index of everything that exists.
  • A source in the registry does not answer every day. Partial coverage is reported, not hidden.
  • An extracted indicator is not automatically confirmed malicious — it is a claim with a source attached.
  • Capture time is attached to every page, so freshness is never inferred from a URL.

Access

Ask us what we would find about you.

Give us a domain and a phrase. We will show you what the corpus holds, how old it is, and what it cannot tell you.

Request accessSee the model family

Coverage figures are updated from the index, not estimated for a marketing page.