Frequently asked questions
Common questions from engineering leaders, security teams, and operations stakeholders evaluating GreenLightz.
Product
What does GreenLightz actually do?
GreenLightz evaluates every commitment your AI agents make — refunds, discounts, delivery promises, warranty extensions — against your company's policies in real-time. It returns a verdict (ALLOW, REQUIRE_APPROVAL, or BLOCK) before the commitment takes effect.
Is this an AI/ML system?
The core evaluation engine is deterministic — rule-based, not ML-based. Same input plus same policy equals the same verdict, every time. This makes verdicts reproducible and auditable. Optional LLM enrichment is available but never on the critical decision path.
What types of commitments can it govern?
Nine built-in commitment types: refunds and credits, discounts and offers, delivery promises, subscription extensions, SLA commitments, data handling pledges, support response commitments, policy exceptions and overrides, and goodwill remedies. Each is evaluated against per-action-type ceilings, aggregate thresholds, and velocity patterns.
Architecture
What if GreenLightz goes down?
The system is fail-closed. Any error — network, storage, or internal — results in BLOCK or REQUIRE_APPROVAL. Your AI agent never gets a false ALLOW due to a system failure.
What's the latency impact?
Designed for sub-millisecond evaluation overhead. P99 under 1 millisecond in our benchmarks. We measure the actual impact in your environment during the pilot.
Does it require external services to work?
No. The core evaluation engine has zero external runtime dependencies. It works fully offline — no LLM calls, no third-party APIs, no network dependencies in the evaluation path.
Can GreenLightz detect when an agent is being pressured or manipulated?
Yes. When your agent platform provides structured conversation signals — such as urgency indicators, manipulation scores, or escalation patterns — GreenLightz evaluates them deterministically alongside the commitment itself. High-risk conversational patterns automatically escalate the verdict. No LLM is used in the decision path — evaluation is fully deterministic and sub-millisecond.
Integration
How does integration work?
One API endpoint. Your AI agent sends a POST request with the action details before executing. GreenLightz responds with a verdict. Any language that speaks HTTP can integrate — no SDK or agent framework lock-in required.
How long does integration take?
Typical integration takes 1-2 days. Most teams wrap the API call in a single helper function. We provide JSON Schema contracts and TypeScript type definitions for code generation.
What if I send PII in the request?
Requests containing PII metadata keys — email, phone, SSN, card numbers, and 7 other patterns — are rejected with HTTP 400 before reaching the evaluation engine. Use opaque identifiers instead.
Security
How is data stored?
All identifiers are SHA-256 hashed with per-tenant HMAC keys before storage — non-reversible. No PII is ever stored. Evidence packets are cryptographically signed and tamper-evident.
How many tests do you have?
Over 7,900 governance test cases across multiple suites — policy evaluation, disclosure governance, operational readiness, integration contracts, and end-to-end simulation. Every test is CI-gated. No deployment ships without a full green run.
Have you done a security audit?
The system has undergone comprehensive internal security hardening — including a 10-plane security audit across 93 hardening iterations with zero unresolved critical or high-severity findings. External third-party audit is in planning. We can discuss our security architecture in detail under NDA.
Pilot
How long is a pilot?
Typically 2-4 weeks. Setup takes 1-2 hours on Day 0. Week 1 runs in shadow mode (evaluate but don't block). Weeks 2-3 run in live mode with real governance. A structured review session closes the pilot.
What does a pilot cost?
Pilot pricing is discussed during the demo call based on your volume and requirements.
Can I test without affecting production?
Yes. Start in shadow mode — GreenLightz evaluates in parallel but doesn't block your agents. You see what would happen without any impact on live operations. Switch to live governance when you're ready.
Still have questions?
30-minute demo. We'll walk through the product, answer your specific questions, and discuss how GreenLightz fits your use case.
Book a Demo