AI safety + compliance for global logistics
AI that classifies goods, drafts customs entries, and quotes shipments operates inside trade law. A hallucinated HTS code or an invented transit promise costs real money on every shipment. EvalGuard grounds those outputs in your tariff and rate data, keeps shipper PII out of third-party models, and caps agent spend at fleet scale.
What ships today
Honest posture, not roadmap promises
Every checked item is in production today. In-progress items are flagged explicitly — no overclaiming, no vapor.
Built for buyer reality
Logistics & supply chain AI use cases we ship for
Customs-entry drafting + HTS classification
AI suggests HTS codes and drafts entry documents from commercial invoices. A wrong classification is a CBP penalty and duty exposure across every repeated shipment — outputs must be grounded and auditable.
EvalGuard features
- Faithfulness scorer grounds classifications in your tariff-ruling corpus
- Citation scorer requires ruling references for every suggested code
- Tamper-evident audit log preserves the basis of every entry for CBP review
- Evidence-bundle export packages classification rationale for brokers
Quoting + rate assistant
Sales teams quote lanes through an AI grounded in rate cards and surcharges. An invented rate or missed surcharge is margin lost on every booked shipment.
EvalGuard features
- Faithfulness scorer checks quoted rates and surcharges against your rate-card corpus
- Output guardrails block quotes outside approved discount authority
- Multi-model routing serves bulk quotes on cheaper models within accuracy thresholds
- Gateway cost ledger attributes AI spend per lane and per customer
Shipment-visibility chatbot
Shippers and consignees query status through a bot with access to tracking + documents. Cross-customer disclosure of routings or commercial terms is a competitive and contractual breach.
EvalGuard features
- PII firewall redacts shipper/consignee identifiers before LLM calls
- Prompt-injection defense: 300+ attack plugins cover cross-account extraction
- Per-route rate limits blunt scraping of network-wide movement data
- Per-project BYOK isolation separates each customer portal
Autonomous ops agents at fleet scale
Agents rebook missed connections, file exceptions, and draft carrier communications around the clock. Without hard cost + action bounds, one looping agent can burn a month's budget overnight.
EvalGuard features
- Per-tenant daily budget caps with a cost-cap kill switch
- Output guardrails hold agent commitments inside approved playbooks
- Audit log records every autonomous action for morning-after review
- Red-team packs probe agent loops for runaway and escalation paths
Wire it in 60 seconds
Wrap your OpenAI client. Ground it in the rate card.
Tariff corpora + discount authority + budget caps are configured once in the EvalGuard control plane. Your code only wraps the client.
import OpenAI from "openai";
import { wrapOpenAI, EvalGuardViolationError } from "@evalguard/openai";
const openai = wrapOpenAI(new OpenAI(), {
apiKey: process.env.EVALGUARD_API_KEY!,
projectId: "customs-entry-drafts",
metadata: { vertical: "logistics", tradeLane: "transpacific" },
blockOnViolation: true, // refuse shipper-data leaks
evalOnResponse: { failOnScore: 0.75 }, // tariff-faithfulness gate
onViolation: (r) => alertTradeCompliance(r.violations),
});
try {
await openai.chat.completions.create({
model: "gpt-4o",
messages: [{ role: "user", content: invoiceExtract }],
});
} catch (err) {
if (err instanceof EvalGuardViolationError) {
// Block recorded in the customs audit trail. Replay via audit ID.
}
}wrapOpenAI for wrapAnthropic.Stack
Six surfaces, one platform
Eval, firewall, red-team, audit, BYOK, dashboard — every surface ships out of the box. No bolt-on vendors, no procurement cycle per capability.
Ready to ship logistics AI you can defend?
Free trial includes the data firewall, faithfulness scorers, and fleet-scale cost caps. Bring a rate card and watch the quote bot stop inventing surcharges.
Apache-2.0 source · SOC 2 evidence engine live · full trust center