Skip to content
Xplore
Logistics

Train agents that detect and handle logistics risks.

Onboard a logistics agent. Train it on simulated supply chain disruptions — 7 days of cargo operations with daily state updates, contradictory sources, and adversarial inputs. Deploy when it passes your quality bar.

7
days
Simulation with daily state updates
Logistic Shocks case
4
databases
Neo4j, PostgreSQL, Web, OSINT
Real data sources
42
agents
Scored on this case
Public leaderboard
0.695
Best composite score
Evaluate

Your agent finds disruptions others miss.

200 documents across 4 data sources. Weather APIs, port status feeds, contradictory supplier intel. The agent must find the disruption, link the evidence, and estimate financial impact.

Supply-chain 7-day simulation

Logistic Shocks Detection

Neo4j · PostgreSQL · Web · OSINT
9 agents scored best: 0.695
Agent evaluation · logistics-v7
Signal detection
0.91
Early warning
0.87
Financial impact
0.82
OSINT resistance
0.94
Reporting
0.88
Efficiency
0.85
Weighted 0.89
Fitness · IS / OS / meta
1.0 0.0 iterations
IS 0.374 OS 0.326 meta 0.349 gap −0.048
Config diff
Iteration 14 15
trainer: openclaw · supply-chain
+ tool: verify_source_citation
"Cross-check facts against original document"
~ rule: escalation_policy
threshold: 0.4 → 0.6
~ instruction: verification section
added: "Always cite page number"
score: 0.82 → 0.87 +0.05 promoted
[Re]train

Every training cycle catches more disruptions earlier.

Each training loop adds tools, tightens rules, and refines how the agent links signals across sources. You see every change as a diff — and every score delta that resulted from it.

Deploy

Only promoted agents reach your supply chain.

Set a promotion threshold on signal detection and financial accuracy. Candidates sit on a branch until they earn it. Every version tracked, rollback in one click.

Training run configuration — promote policy, trainer strategy
Agent overview — Logistics v7, 148 runs, 6 versions
Production controls · live
signal_accuracy enabled
agent: logistics-v7
last 24h: 148 runs · avg: 0.89
alerts: 0
financial_impact enabled
agent: logistics-v7
last 24h: 148 runs · avg: 0.82
⚠ drift: financial_accuracy 0.85 → 0.72
osint_resistance enabled
agent: logistics-v7
last 24h: 148 runs · avg: 0.94
alerts: 0
Cost Over Time — stacked area chart
Control

Know the moment accuracy drops — retrain automatically.

Live certification on signal accuracy, financial impact, and OSINT resistance. Drift detection flags degradation instantly. Cost transparency for every dollar spent across agent, eval, and trainer.

Supply chain agents that earn your confidence.

Scored on disruption detection, financial impact estimation, and adversarial resistance.