White Force Agent
Exercise control, training variant only.
- Doctrine alignment
- Aligned to NATO exercise planning and Joint Warfare Centre practice.
- Deployment
- Training variant only; not fielded on operational systems.
- Underlying products
- SynapseMesh™VerityGridAI™
This agent runs on the training variant of the platform. It is compiled out of operational builds and is never fielded on operational systems.
Purpose
Training exercises depend on consistent, fair exercise control, yet human controllers struggle to apply adjudication rules uniformly across long, complex events. SC-WHT-01 acts as exercise control for training and experimentation: it adjudicates injects, tracks training objective coverage and maintains the exercise narrative. Adjudication rules are declared before STARTEX and applied consistently, with every ruling logged alongside its rule and rationale, which ends most post-exercise disputes before they start. The agent is compiled out of operational builds, not merely disabled, so it cannot be accidentally enabled on a live system.
Architecture
The exercise specification, training objectives and the live stream of player actions and reports drive an exercise-control core that adjudicates injects, tracks objective coverage and maintains the exercise narrative. Adjudication rules are declared before STARTEX and applied consistently throughout, and every adjudication is logged with its rule and rationale, which ends most post-exercise disputes before they start. The agent runs on the training variant of the platform only and is compiled out of operational builds, not merely disabled.
Hover or focus a band to see what it guarantees.
Inputs and outputs
- IN-01Exercise specification
- IN-02Training objectives
- IN-03Player actions and reports
- OUT-01Inject adjudication
- OUT-02Exercise control logs
- OUT-03Training objective coverage
Position in the mesh
- Consumes
- Publishes
What it does not do
It is never fielded on operational systems. There is no operational configuration that includes it.
The same four gates as every agent
An unregistered or unapproved model tag is refused before inference, not flagged afterwards.
Every action is scoped to a declared role. Scope is checked at the engine, not in the prompt.
New or changed models run under supervision until their evaluation evidence is accepted.
Irreversible decisions require a signed human authorisation recorded with the decision.
The agents differ in what they reason about. They do not differ in what they are allowed to do.
Deployment
In-country, accredited cloud, under your jurisdiction and operational control.
Designed for deployment up to NATO SECRET. Accreditation is per deployment.
Runs disconnected. Inference, agents and the decision record are all local.
Training variant only; not fielded on operational systems.
Ready to see SynapseCommand® in action?
Briefings available under NDA. Classified briefings arranged through cleared channels.