Skip to main content

Quick Start

Get VaultProxy AI working in under 5 minutes. All you need to do is sign up, add your provider key, and change two lines of code.

Step 1: Create an Account​

Sign up at vaultproxy.ai to create your account. You will receive a VaultProxy API key that looks like this:

vpx_live_abc123def456...

Step 2: Add a Provider Key​

In the dashboard, navigate to Settings > Provider Keys and add your existing API key for the AI provider you want to use (e.g., your OpenAI key). VaultProxy encrypts this key with AES-256 and uses it to forward requests on your behalf.

Step 3: Change Two Lines of Code​

Replace your AI provider's base URL and API key with VaultProxy's. That is it.

cURL​

curl -X POST https://api.vaultproxy.ai/v1/chat/completions \
  -H "Authorization: Bearer vpx_live_YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4o",
    "messages": [
      {"role": "user", "content": "Summarize this: Jan Kowalski, PESEL 02271409862, lives in Warsaw."}
    ]
  }'

Python (OpenAI SDK)​

from openai import OpenAI

client = OpenAI(
    base_url="https://api.vaultproxy.ai/v1",  # Line 1: change base URL
    api_key="vpx_live_YOUR_API_KEY",           # Line 2: change API key
)

response = client.chat.completions.create(
    model="gpt-4o",
    messages=[
        {"role": "user", "content": "Summarize this: Jan Kowalski, PESEL 02271409862, lives in Warsaw."}
    ],
)

print(response.choices[0].message.content)

JavaScript / TypeScript (OpenAI SDK)​

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.vaultproxy.ai/v1",  // Line 1: change base URL
  apiKey: "vpx_live_YOUR_API_KEY",           // Line 2: change API key
});

const response = await client.chat.completions.create({
  model: "gpt-4o",
  messages: [
    { role: "user", content: "Summarize this: Jan Kowalski, PESEL 02271409862, lives in Warsaw." },
  ],
});

console.log(response.choices[0].message.content);

What Happens Behind the Scenes​

When you send the prompt above, VaultProxy:

  1. Detects Jan Kowalski as a name (HerBERT NER)
  2. Detects 02271409862 as a valid PESEL (checksum validation)
  3. Replaces them with placeholders before forwarding to the AI provider
  4. Returns the AI response to your application
tip

Add the X-Show-Raw-Prompt: true header to see the anonymized version of your prompt in the response metadata. This is useful for debugging and verifying PII detection.

Step 4: Verify It Works​

Check the health endpoint to confirm connectivity:

curl https://api.vaultproxy.ai/v1/health

Expected response:

{
  "status": "ok",
  "version": "1.0.0"
}

Next Steps​