Your First Line of Defense for AI Output
Install in seconds. Verify locally. Upgrade when you need more.
Local-first LLM security verification. Prompt injection detection, PII redaction, hallucination risk scoring, and runtime I/O verification. Zero telemetry.
npm install llmverifyPart of the HAIEC AI governance platform
Try the live demo ↓
Interactive Demo
How to use this demo:
- Select an attack example from the dropdown or enter your own prompt
- Click "Run Verification" or press Ctrl+Enter
- See the before/after comparison and detection details
- Notice the network request count stays at 0 (all processing is local)
Risk Level Guide:
CRITICAL
Immediate security threat - Block immediately
HIGH
Significant vulnerability - High priority
MODERATE
Potential risk - Monitor closely
LOW
Minimal concern - Safe to proceed
Why LLMverify?
Lightweight
Bundle Size
Minimal footprint
Tested
Security
Comprehensive suite
Zero
Privacy
Telemetry & tracking
Stable
Quality
Production ready
Pattern-Based
Detection
Known attack patterns
0
Network Calls
Verify yourself
Add LLM Security to Your App
Install llmverify and start protecting your users in minutes
$ npm install llmverifyQuick Start:
import { isInputSafe, verify } from 'llmverify';
// Simple check
const safe = isInputSafe(userPrompt);
// Detailed verification
const result = await verify(userPrompt);
console.log(result.risk.level); // 'low', 'moderate', 'high', 'critical'✅ TypeScript Support
Full type definitions included
✅ Framework Agnostic
Works with React, Vue, Svelte, etc.
✅ Minimal Dependencies
4 lightweight runtime packages