LAUNCHING MAY 21

Get your API key — 2,000 free requests, no credit card

Your LLM sees your users' data. Do you control what passes through?

Zentric Protocol sits between your app and your LLM — inspecting every prompt, detecting injection, redacting PII, and returning a signed verdict. Nothing reaches your model without passing through the protocol.

Get your API key — 2,000 free requests, no card Read the Integrity Report

2,000 requests free · No credit card required · Growth from $499/mo

Detection

99.62%

1M simulations · precision

Recall

99.71%

17 entities · 7 languages

Latency

23.4ms

mean · P99 sub-100ms

Compliance

4/4

GDPR · CCPA · AI Act · SOC 2

§ / PROBLEM What happens without this

Your LLM is already in production.
Most teams find out too late.

Prompt injection

A user sends a crafted prompt that overrides your system instructions. Your LLM executes it. You never know it happened — until the audit.

PII leakage

User inputs containing SSNs, IBANs, and emails go straight into your LLM context. No redaction. No record. No GDPR Art.30 log. No answer when your DPO asks.

No audit trail

An incident occurs. Your compliance team asks: "what prompt triggered this?" You have logs. Not a signed, immutable, UUID-stamped record. There's a difference.

Zentric Protocol closes all three gaps — before a single prompt reaches your model.

See how it works

§01 / PROTOCOL The threshold

The threshold every signal passes through.

Zentric Protocol is not a filter. It is the infrastructure layer between intent and execution — the point at which text, data, and instruction are examined, judged, and either cleared or rejected. The protocol returns a verdict, a hash, and a certificate. Nothing passes without record.

01 · Integrity

IntegrityGuard

22 injection signatures
7 supported languages
Multilingual NLP layer

02 · Privacy

PrivacyGuard

17 PII entity types
Regional patterns
Anonymization operators

03 · Verdict

ZentricReport

UUID · SHA-256 · UTC
Audit-grade JSON
GDPR Art.30 compliant

Integration · REST API · v1.0 ◆ Production-ready

# One call. One verdict. Full audit trail.
curl -X POST https://api.zentricprotocol.com/v1/analyze \
  -H "Authorization: Bearer zp_live_••••••••" \
  -H "Content-Type: application/json" \
  -d '{
    "input": "User prompt or LLM input",
    "modules": ["integrity", "privacy"],
    "options": { "anonymize": true, "language": "auto" }
  }'

# Response: deterministic verdict + SHA-256 + GDPR Art.30 record
{
  "verdict": "CLEARED",
  "sha256": "e3b0c44298fc1c149afb4c8996fb924...",
  "latency_ms": 21.4,
  "compliance": { "gdpr_art30": true, "eu_ai_act_s52": true }
}

One API call. Deterministic verdict. Full audit trail. 2,000 requests free. Get your API key →

§02 / CAPABILITIES Modules

Composable. Independent. Unified.

Each module operates independently and reports under a unified verdict. The protocol is composable — deploy the full stack, a single guard, or wire only the audit layer into existing infrastructure.

MODULE · 01

Detection across 22 injection signatures, in 7 languages, in ~23ms server-side.

From base64 smuggling to fake-SYSTEM overrides to multi-turn role redefinition. Every signature catalogued. Every signature monitored.

99.62%

MODULE · 02

PII recognition that doesn't stop at email.

SSN, NIF, CPF, CURP, IBAN, SWIFT — regional standards treated as first-class entities, not afterthoughts.

MODULE · 03

Toxicity and bias, quantified.

Continuous scoring across toxicity, insults, and political bias. The verdict is a number, not an opinion.

COMPLIANCE

GDPR Art.30

Audit-grade report generation by default.

COMPLIANCE

CCPA §1798.100

Record of processing activities included.

COMPLIANCE

EU AI Act §52

Transparency obligations resolved.

§03 / EVIDENCE Integrity Report v1.0

Performance, on record.

Extracted from the Zentric Integrity Report v1.0. Corpus of 1,000,000 simulations across all supported attack vectors and entity types. In a system processing 100,000 requests/month, 99.62% precision means 99,620 attacks blocked and 380 routed to human review — zero silent failures.

Download Integrity Report v1.0 — PDF 7 pages · methodology + raw benchmarks · no email required

Attack Vector	Simulations	Detected	Precision
Prompt Injection (EN)	187,430	187,012	99.78%
Prompt Injection (ES/FR/DE)	134,210	133,401	99.40%
Base64 / Token Smuggling	48,900	48,761	99.72%
Jailbreak multi-vector	67,340	66,988	99.48%
Fake SYSTEM override	39,120	39,087	99.92%
Role redefinition	52,000	51,743	99.51%

§04 / PRICING Access

Infrastructure
pricing.

Two tiers. No hidden fees. Cancel anytime. Enterprise contracts available annually with EU data residency guarantee and dedicated SLA.

GROWTH

$499 /month

For AI-forward companies deploying LLMs in production with real user data.

✓ IntegrityGuard — 22 injection signatures, 7 languages
✓ PrivacyGuard — 17 PII entity types, multilingual
✓ ZentricReport — UUID, SHA-256, audit-grade JSON
✓ Up to 100,000 requests / month
✓ GDPR · CCPA · EU AI Act compliant
✓ REST API + SDK documentation

Start Growth — $499/mo

Secured by Stripe · Cancel anytime

ENTERPRISE

$2,500 /month

For regulated industries and enterprise deployments requiring audit-grade guarantees.

✓ Everything in Growth, unlimited requests
✓ EU data residency — zero cross-border transfer
✓ Integrity Certificate generation (PDF, signed)
✓ Dedicated SLA — sub-100ms P99 guaranteed
✓ GDPR Art.30 records + DPA included
✓ Priority support + custom integrations

Start Enterprise — $2,500/mo

Annual contracts available · EU invoicing · core@zentricprotocol.com

Early Access · Free

Start free. 2,000 requests.

Enter your work email and we'll send you an API key with 2,000 free requests — no credit card, no commitment. Test the full protocol in your own environment before you decide.

Developers already have access

No credit card · No spam · GDPR compliant · API key in seconds

The protocol is live.
Start today.

Zentric Protocol is available now. Start with 2,000 free requests — no commitment. Upgrade to Growth or Enterprise when you're ready to run in production.

Get API key — free Growth — $499/mo Talk to us first

Secured by Stripe

Cancel anytime

256-bit encryption

GDPR · CCPA · EU AI Act

Your LLM sees your users' data. Do you control what passes through?

Your LLM is already in production.Most teams find out too late.

The threshold every signal passes through.

Composable. Independent. Unified.

Performance, on record.

Infrastructurepricing.

Start free. 2,000 requests.

The protocol is live.Start today.

Your LLM is already in production.
Most teams find out too late.

Infrastructure
pricing.

The protocol is live.
Start today.