Enforcement Intelligence Platform

Automated Enforcement Intelligence
for Online ENDS Retailers

Forensic compliance auditing across 2,600+ online vape stores. FDA authorization analysis, payment processor identification, interstate shipping probes, and evidence-grade violation reports — built for law enforcement.

2,600+Stores Audited
13,000+FDA-Unauthorized Products Identified
33Payment Processors Tracked
10Restricted States Probed
50Jurisdictions Covered

Six forensic tools.
One evidence-grade report.

FDA Authorization Analysis

Every product checked against all 47 FDA-authorized ENDS products. Four-tier classification: MDO brands, Warning Letter brands, generic unauthorized, and confirmed authorized. Youth-appeal flavors, synthetic nicotine, and high-puff disposables flagged. Each finding cites 21 U.S.C. § 387j, § 387b(6)(A), and § 331(a).

Payment Processor Intelligence

33 processors detected through five parallel methods: network interception, iframe scanning, JS global evaluation, DOM text analysis, and unknown request capture. Each processor is mapped with the specific Terms of Service section violated.

Interstate Shipping Probes

End-to-end checkout verification across 10 restricted states. Two-stage: cart API zone testing then real Chromium checkout walk-through. Verdicts: END_TO_END_CONFIRMED, ZONE_OPEN, CHECKOUT_BLOCKED, or ZONE_CLOSED — with full PACT Act exposure assessment.

Evidence-Grade Reports

Formal HTML/PDF enforcement reports with statute citations, product evidence tables, screenshots, PACT Act exposure analysis, and recommended actions — structured for cease-and-desist letters, CIDs, legal filings, and payment processor referrals.

Four-Phase Forensic Pipeline

Phase 1: AI reconnaissance and hypothesis formation. Phase 2: Seven concurrent evidence collection tools. Phase 3: AI validation with confidence scoring. Phase 4: Formal enforcement narrative generation with uncertainty acknowledgement.

Multi-Jurisdiction Access

Role-based secure access with jurisdiction-level views across all 50 states and DC. State-by-state aggregate scores, retailer concentration maps, and triage queues for supervisory review. Each finding includes confidence scores and documented evidence gaps.

Built to the standard
enforcement actions require

47
FDA-Authorized ENDS Products

Every product on every audited store is classified against the complete FDA-authorized set — not estimated, not sampled.

§ 387j
Statute-Cited Findings

Each violation references the specific FD&C Act provision violated. Reports are structured for direct use in legal proceedings.

4-Phase
AI-Validated Pipeline

Mercury AI forms hypotheses, tools collect evidence, AI validates against hypotheses, then generates the narrative — with uncertainty formally documented.

100%
Evidence Retained

Screenshots, HTML artifacts, network captures, and product data are preserved per audit. Nothing is inferred without a citable source.

What an enforcement report
actually looks like

Each audit produces a formal enforcement report with statute citations, product evidence tables, payment processor findings, and interstate shipping verdicts. The example below reflects the standard output for a high-risk ENDS retailer.

EXTREME RISKSample Enforcement Target
585Unauthorized Products
5Violation Categories
END_TO_ENDShipping Confirmed
21 U.S.C. § 387j585 PMTA-unauthorized ENDS products sold without FDA marketing authorization
PACT Act § 2(a)Interstate delivery shipping confirmed via end-to-end checkout probe to restricted state
21 U.S.C. § 387b(6)(A)Youth-appeal products (candy/fruit/dessert flavors) actively listed and purchasable

From AI reconnaissance to enforcement-ready report

01

AI Preliminary Investigation

Mercury AI reconnaissance: store type identification, MDO and Warning Letter brand detection, risk hypothesis formation, and statute-linked suspected violations — before any evidence tools run.

02

Forensic Evidence Collection

Seven concurrent tools: catalog extraction, browser crawling, age-gate detection, policy scanning, interstate shipping probes across 10 restricted states, and payment processor identification via network interception.

03

AI Validation & Confidence Scoring

Evidence tested against initial hypotheses. Each violation is CONFIRMED, UPGRADED, DOWNGRADED, REJECTED, or flagged UNVERIFIED. New violations are surfaced. Evidence gaps are formally documented.

04

Enforcement Report Generation

Statute citations verified, evidence linkage established, recommended actions organized by receiving agency (FDA, FTC, payment processors, AG offices), and uncertainty formally acknowledged in writing.

Authorized access

Access is limited to authorized enforcement personnel. Contact hello@mercurylabs.ca to request access or report provisioning.

Sign In to Portal