LIVE — measured from production APIs

HERO

The HelmLegal fraud-fighting engine, projected publicly. Every number below is fetched live — not a screenshot, not a pitch deck. These systems read public court records, verify evidence integrity, and gate every claim before publication.

--
Fraud Patterns
--
Cases Tracked
--
Entities Mapped
--
Docket Entries
--
Public Claims
--
Gate Pass Rate
--
Dockets Mirrored

The models in the fight

PATTERN-ENGINELIVE
Classifies fraud patterns across docketed events — mail fraud, conversion, procedural abuse, securities fraud and 30+ other typologies. Feeds Fraud Radar and the public record feed.
TRUTH-ENGINELIVE
Contradiction detection across sworn filings, testimony, and exhibits. Where two representations cannot both be true, the contradiction is catalogued — see the public claims register and verification ledger.
DOCKET-MIRRORLIVE
Mirrors court dockets into verifiable, timestamped rows. Sealed and privileged entries are marked but never published. Live at /docket-mirror.
EVIDENCE-GATELIVE
Hash-first artifact verification and chain-of-custody checks. Nothing reaches a public surface without passing the gate — measured pass rate shown above. Methodology: /methodology.
READINESS-TIERINTERNAL
Gate severity analytics, staleness detection, and package readiness scoring for counsel review. Auth-gated by design — internal operations only.
TRANSPARENCY-LEDGERLIVE
Every public claim carries a state — SUPPORTED, PARTIAL, CONTRADICTED — plus corrections and open gaps. See /transparency, /gap-register, and /drift.

Detected pattern classes (live)

Loading…

Measured live from api.helmlegal.net/api/v1/fraud/stats and /api/v1/transparency-metrics. Last rendered client-side; view source to verify.