Trust and identity

Local identity and endpoint security.

TrustID v3 combines behavioural signals, system checks, a trust score and responses defined by the operator’s policy. Processing takes place on the protected endpoint.

• Windows architecture with eleven detection layers.
• CCD engine with eight specialised agents and case memory.
• Local collection, correlation and decisions.

Layered protection with a local trust score.

The Windows v1.5 edition correlates endpoint signals and applies responses according to its rules. It is presented separately from v3 and does not include v3’s CCD engine.

• Weighted trust scoring.
• Local monitoring and responses.
• Rules matched to the deployment profile.

The hospital edition of the TrustID v3 engine.

TrustID Hospital brings a healthcare layer to the TrustID Core v3 engine: DICOM and RIS signals, clinical identities, role-based access control, medical behaviour patterns and a signed audit chain. Processing stays on the institution’s infrastructure.

• Detection of abnormal patient-record access, bulk exports and unjustified break-glass access.
• Hybrid RSA-4096 and ML-DSA-65 audit signatures; AES-256-GCM encryption at rest.
• No outbound traffic in the default configuration; PACS, RIS, directory and SIEM connectors are enabled by the institution.

Checking whether content is authentic, directly on the phone.

Elyne Verify is a fully local Android content-verification app: a dedicated detector scores AI-generated images and videos, and an embedded multimodal model explains the verdict in plain language. “Send to Elyne” works from other applications through sharing.

• ONNX detector and arm64 Gemma engine run on the device, with no data sent out.
• Credibility analysis of news articles is announced, not delivered.
• Born from TrustID’s mobile strand, refocused on verification.

Local defence alongside agentic control.

Elyne Cyber observes system signals, checks them against its policy and applies bounded responses. Models can assist analysis; action authorisation remains deterministic. The service is distinct from EAS Defense.

• Local collection and correlation.
• Responses defined by operator rules.
• TrustID signals when that integration is configured.

Prepare SOC and CERT teams for adversarial scenarios.

Cyber Red provides real attack scenarios within an authorised laboratory. It helps exercise defence teams and observe the tested environment’s responses.

• Real actions within a defined scope.
• Exercises for SOC and CERT teams.
• Effect observation and outcome review.

An exercise adversary combining attack and defence.

Cyber Hybrid is the training variant combining offensive and defensive behaviour. It targets response exercises: blocking, containment and regaining the initiative in an authorised environment.

• Offensive and defensive behaviour in one exercise.
• Laboratory scope agreed with the operator.
• Outcomes assessed for each scenario.