OSINT Research

Public-Source Research Your Reader Can Verify.

Markets, competitors, contested numbers: ask the question and @vustSearchBot synthesizes an answer from the live public web with numbered, openable sources. Escalate through Research and the agent-driven Deep Report, and cross-check high-stakes claims across three models. Public data only — this is a research aid, not surveillance tooling.

From a single lookup to an agent-compiled report on Pro — exact price shown before you confirm.Public web · verifiable citations
Openable citations on every answerLookup → Research → Deep ReportThree-model cross-check

Honest scope

The research half of OSINT — without the tradecraft half

The tool synthesizes over the live public web and always shows its sources. What it doesn't have and won't: breach-database lookups, photo geolocation, face search, social-graph mapping or any people-search features. If information isn't publicly indexed, it isn't here — and the answer won't pretend it is.

Research in @vustSearchBot · cross-check in @vustbot.
Specimens

See the difference

Three research questions — and what comes back for each.

Competitor footprint brief

Your question

"What has [competitor] publicly shipped and priced in the last year — and what did coverage say about adoption?"

What comes back

A synthesized brief drawn from their announcements, pricing pages, press coverage and public filings — each claim carrying a numbered, openable link. The Research tier widens it with multi-angle sub-searches; sources come with domains and snippets so you can vet before you cite.

Market-size sanity check

Your question

"Roughly how big is the [niche] market, and whose estimates disagree?"

What comes back

The public estimates that exist, who published each, and where they diverge — stated as a spread with sources, not one confident number. If sources disagree on a key figure, the answer says so instead of papering over it.

The claim you can't afford to get wrong

Your finding

A single striking stat that's about to anchor your deck's argument.

The cross-check

Run it past the Council of Sages in @vustbot: Claude, Grok and Gemini answer independently, and an arbiter names where they disagree. Three models agreeing doesn't make a claim true — but three models disagreeing is a flag worth seeing before your audience finds it.
Practical use cases

Public-source research with citations — market, competitor, claim

Market researchers
Size a niche or map who competes in it before committing budget
Ask in plain language and get a synthesized answer with numbered, openable source links — press releases, filings, coverage — so every number in your brief has an origin you can show.
Competitive analysts
Track what a competitor shipped, priced or announced from their public footprint
The Research tier runs a wider multi-angle search with a dedicated source display; the Deep Report tier sends an agent to browse, follow leads and compile a structured report.
Claim verifiers
A stat is about to go into your deck and being wrong is expensive
Cross-check it with the Council of Sages — Claude, Grok and Gemini answer independently and an arbiter names the disagreements — before the claim becomes yours.
How it works01–03

The OSINT-style workflow, step by step

  1. 01

    Ask the research question

    "Who are the main players in X and how do they price?" — on the Research tier @vustSearchBot synthesizes an answer from live public web sources with numbered citations you can open.

  2. 02

    Go deeper where it matters

    Escalate to Research (multi-angle sub-searches, source snippets) or the Pro-tier Deep Report — an agent that browses pages, follows leads and returns a structured report with a full source list.

  3. 03

    Cross-check before you commit

    Contested or high-stakes findings go to the Council of Sages in @vustbot: three independent models plus an arbiter that surfaces where they disagree, instead of one model grading itself.

Same tool · in Telegram@vustSearchBot

Sourced research in Telegram

Ask @vustSearchBot a market or competitor question. On the Research tier the answer carries numbered markers and openable sources; Deep Report compiles an agent-driven report on Pro. Tier prices are fixed and known before you send.

Open in Telegram
Quality & trust

Honest scope — research aid, not surveillance tooling

Public sources only, and only what search reaches

Answers synthesize from the live public web — no scraping behind logins, no data brokers, no dark-web access, no historical archives beyond what public pages state. If the information isn't publicly indexed, this tool won't have it, and the answer won't pretend otherwise.

Not a people-tracking tool

This is for market, competitor and claim research. It is not built for locating, profiling or de-anonymizing individuals — requests in that direction are the wrong use of the tool, and classic OSINT tradecraft (breach lookups, image geolocation, social-graph mapping) is simply not what it does.

Citations are the product

Every answer carries openable source links; the deeper tiers show domains and snippets. Findings you can't trace to a source you can open shouldn't go in your report — and here they don't have to.

FAQ

Frequently asked questions

Is this a real OSINT tool?

It's the research half of OSINT — synthesis over open sources with verifiable citations — without the tradecraft half. It searches the live public web and returns answers whose every claim links to an openable source; the deeper tiers browse and compile structured reports. What it deliberately is not: breach-database lookups, image geolocation, social-graph mapping or any people-tracking workflow.

What sources does it actually reach?

The live public web as a search engine reaches it: news, company pages, filings, documentation, public social posts, coverage. It does not log into anything, buy data-broker records, access the dark web, or see content behind paywalls and logins. If a fact isn't publicly indexed, the honest answer is that it can't be found here — and that's the answer you'll get.

Can I use it to research a person?

That's the wrong use of this tool, and we say so on the page. It's built for market, competitor and claim research over public sources. It has no people-search features — no breach lookups, no face or image search, no de-anonymization — and requests aimed at locating or profiling individuals are outside what it does.

How is this different from just using the /search page?

Same engine, different job. /search is the general tour of sourced answers at four depths. This page maps a specific workflow onto those pieces: question → sourced brief → multi-angle Research or an agent-compiled Deep Report → Council cross-check for the claims that carry weight. If you just need one answer with sources, /search covers it; if you're assembling a research brief someone will challenge, this is the workflow.

What does the escalation ladder cost?

Quick, Standard and Research show their exact Balance price before confirmation. Deep Report — the agent that browses, follows leads and compiles a structured report — comes with the Pro subscription under live cost and safety guardrails, alongside the Council of Sages.

How do I know the findings aren't hallucinated?

Two mechanisms, honestly framed. First, citations: every answer links to openable sources, so a claim you can't trace doesn't have to be trusted. Second, disagreement: the Council of Sages runs three independent models and surfaces where they split. Neither makes an answer automatically true — they make it checkable, which is the actual standard a research brief has to meet.

Ready when you are

Research briefs your reader can verify.

Public-source answers with openable citations, multi-angle deep dives, and a three-model cross-check for the claims that matter — in Telegram on one Balance.