Lessons-Learned & Assessment Agent
Predicted vs. actual, auditable artefacts.
- Doctrinal basis
- NATO Lessons Learned Capability framework.
- Deployment
- Long-term archival on tamper-evident ledger.
- Underlying products
- CyberFort Blockchain™VerityGridAI™
Purpose
After-action reviews frequently rely on reconstructed narratives and disputed recollections, because the original predictions were never recorded or time-stamped. SC-LES-01 compares predicted to actual outcomes by reading the sealed decision record. Because every prediction is retained in the hash-chained record and, in assured deployments, ML-DSA-signed when produced, variance analysis is an audit rather than a reconstruction. It produces trend analyses and cryptographic audit packages that a third party can verify without trusting the platform, supporting the NATO Lessons Learned Capability framework with evidence that cannot be quietly rewritten.
Doctrinal basis
NATO Lessons Learned Capability framework.
Architecture
The agent works entirely from the sealed record: decision logs, operational logs and sensor truth data. Because every prediction made by the other agents was signed and anchored at the time it was made, predicted versus actual comparison is an audit, not a reconstruction. The core computes variance reports and trend analyses across operations and exercises, and packages findings as cryptographic audit packages that a third party can verify without trusting the platform. Long-term archival sits on the tamper-evident ledger, aligned to the NATO Lessons Learned Capability framework.
Hover or focus a band to see what it guarantees.
Inputs and outputs
- IN-01Operational logs
- IN-02Sensor truth data
- IN-03Decision logs
- OUT-01Variance reports
- OUT-02Trend analyses
- OUT-03Cryptographic audit packages
Position in the mesh
- Consumes
- Publishes
- All agents (decision records)
- SC-WHT-01White Force Agent
What it does not do
It cannot edit history. The ledger it reads is append-only for the agent as well.
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.
Ready to see SynapseCommand® in action?
Briefings available under NDA. Classified briefings arranged through cleared channels.