api-request.ts
1 import { ProxyClient } from '@proxy/sdk';
2
3 const client = new ProxyClient({
4 apiKey: process.env.PROXY_API_KEY,
5 });
6
7 const response = await client.chat({
8 model: 'openai/gpt-4o',
9 messages: [
10 { role: 'system', content: 'You are helpful.' },
11 { role: 'user', content: 'Hello!' }
12 ],
13 temperature: 0.7,
14 max_tokens: 1024,
15 });
16
17 console.log(response.content);
TypeScript | UTF-8
Connected
Model switches in real-time
50+
AI Models
99.9%
Uptime
Minimum
Added Latency
Scale

Security, observability & control

Everything you need to ship safely on day one — without building your own gateway or maintaining custom scripts.

Key protection

Store provider keys on the server, never in the client. Rotate and revoke instantly from the dashboard.

Rate limiting

Per-user and per-endpoint limits stop abuse before it hits your provider bill.

Realtime logs

Inspect requests, responses and errors in one place. Filter by user, endpoint or provider.

Model overrides

Switch models or providers without shipping a new app build. Useful for A/B tests.

Usage alerts

Get notified when spend or token usage crosses thresholds you set.

Team ready

Invite teammates, share services and manage production vs. staging keys.

Works with your favorite AI providers

Bring your own keys for any provider — or route everything through a single ProxyGuard key.

OpenAI Anthropic Mistral Gemini Together Groq

Integrate in minutes

Point your existing OpenAI, Anthropic or other SDKs at the ProxyGuard endpoint. No new SDKs or custom auth flows required.

  • • Keep your secret keys on the server
  • • Issue per-customer proxy keys from your dashboard
  • • Log, throttle and override requests without touching client code
# Example: Node.js

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.yourproxy.com/v1",
  apiKey: process.env.PROXY_KEY,
});

const completion = await client.chat.completions.create({
  model: "openai/gpt-4.1-mini",
  messages: [{ role: "user", content: "Hello from ProxyGuard!" }],
});

FAQs

Still curious? Here are some answers to common questions.

Do you store provider API keys?

Yes. Provider keys are encrypted at rest and never exposed to client apps. You can rotate them from the dashboard at any time.

Can I bring my own OpenAI/Anthropic keys?

Absolutely. ProxyGuard is a "bring your own key" proxy — we just sit in the middle to add security and observability.

How much work is it to integrate?

Most teams can drop in the proxy URL and env var in under an hour. No infra to maintain.