Now accepting founding members

LLM guardrails
that just work

PII detection, toxicity scoring, prompt injection defense. One API call.

Join the Waitlist
POST /v1/guard
// Input
"My SSN is 123-45-6789"
// Output
"My SSN is [REDACTED]"
✓ PII detected ✓ Toxicity: 0.02 ✓ Injection: clean

Enterprise-grade AI safety

🔐

PII Detection

Credit cards, SSN, IBAN, emails, phone numbers. Powered by Microsoft Presidio + spaCy.

☠️

Toxicity Scoring

5 categories: toxic, severe_toxic, obscene, threat, insult. Real-time Detoxify analysis.

🛡️

Prompt Injection Defense

11 regex patterns + heuristics. Catch attempts to override system prompts.

🚫

Jailbreak Detection

Identify attempts to bypass AI safety guidelines. Pattern + semantic analysis.

📋

Schema Validation

Ensure LLM outputs match your expected structure. Pydantic-powered validation.

How it works

1

Sign up

Get your API key in seconds. Free tier included.

2

Guard your LLM

POST /v1/guard

POST /v1/guard with your text. Choose which guards to run.

3

Ship with confidence

Block, flag, or redact. Configurable actions per guard.

🛡️

Founding Member Offer

Join the waitlist before launch and lock in

35% off forever

This discount applies to any paid plan and never expires. Only for early supporters.

Simple, transparent pricing

Founding member prices shown. Lock them in by joining the waitlist.

Free

$0 /mo

For trying things out

  • ✓ 5,000 checks/mo
  • ✓ Basic guards
  • ✓ Community support

Starter

$49 $32 /mo

For small projects

  • ✓ 50,000 checks/mo
  • ✓ All guards
  • ✓ Email support
  • ✓ Custom thresholds
POPULAR

Pro

$99 $64 /mo

For growing products

  • ✓ 500,000 checks/mo
  • ✓ All guards + custom thresholds
  • ✓ Priority support
  • ✓ Webhooks & analytics

Business

$199 $129 /mo

For teams at scale

  • ✓ Unlimited checks
  • ✓ All features + SLA
  • ✓ Dedicated support
  • ✓ On-prem available

Get early access

Join the waitlist and be the first to know when GuardPost launches.