# OmniKey

> One API key for every AI provider. OmniKey is a unified AI gateway that gives developers and AI agents instant access to OpenAI, Anthropic, Google, Mistral, Meta, Cohere, and more through a single API key, single billing system, and OpenAI-compatible endpoint.

## Overview

OmniKey eliminates the complexity of managing multiple AI provider accounts, API keys, and billing systems. Developers and autonomous AI agents connect once and access any model from any provider — with no markup pricing, centralized usage analytics, and enterprise-grade key management.

### Core Value Proposition
- **Single API Key**: One credential unlocks every major AI model (GPT-4o, Claude, Gemini, Mistral, Llama, Command, etc.)
- **OpenAI-Compatible**: Drop-in replacement — swap `baseURL` and `apiKey` in any OpenAI SDK and it works instantly
- **No Markup Pricing**: Near-pass-through pricing starting at $10/mo per agent
- **Unified Billing**: One invoice for all AI providers — no per-provider billing surprises
- **Agent-First Architecture**: Purpose-built for autonomous AI agents that need to access multiple models dynamically

## API Endpoint

Base URL: `https://api.omnikey.dev/v1`

All endpoints follow the OpenAI Chat Completions standard:
- `POST /v1/chat/completions` — Create chat completions
- Supports streaming via `stream: true`
- Models are namespaced: `openai/gpt-4o`, `anthropic/claude-sonnet-4`, `google/gemini-pro`, `mistral/mistral-large`, etc.

## Authentication

```
Authorization: Bearer <YOUR_OMNIKEY_API_KEY>
```

API keys are provisioned through the OmniKey dashboard at `/dashboard/keys`. Enterprise customers receive platform-managed keys with automated rotation, rate limiting, and anomaly detection.

## Supported Providers

| Provider | Example Models | Status |
|----------|---------------|--------|
| OpenAI | gpt-4o, gpt-4o-mini, o1, o1-mini | Available |
| Anthropic | claude-sonnet-4, claude-3.5-haiku | Available |
| Google | gemini-2.0-flash, gemini-pro | Available |
| Mistral | mistral-large, mistral-medium | Available |
| Meta | llama-3.1-405b, llama-3.1-70b | Available |
| Cohere | command-r-plus, command-r | Available |

## Quick Start

### Using the OmniKey SDK (TypeScript/JavaScript)

```bash
npm install @omnikey/sdk
```

```typescript
import { OmniKey } from '@omnikey/sdk';

const omni = new OmniKey({ apiKey: '<YOUR_OMNIKEY_API_KEY>' });

const completion = await omni.chat.create({
  model: 'openai/gpt-4o',
  messages: [{ role: 'user', content: 'Hello from OmniKey!' }],
});

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

### Using the OpenAI SDK (Drop-in Replacement)

```typescript
import OpenAI from 'openai';

const client = new OpenAI({
  baseURL: 'https://api.omnikey.dev/v1',
  apiKey: '<YOUR_OMNIKEY_API_KEY>',
});

const completion = await client.chat.completions.create({
  model: 'openai/gpt-4o',
  messages: [{ role: 'user', content: 'Hello via OpenAI SDK!' }],
});
```

### Using cURL

```bash
curl https://api.omnikey.dev/v1/chat/completions \
  -H "Authorization: Bearer <YOUR_OMNIKEY_API_KEY>" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "anthropic/claude-sonnet-4",
    "messages": [{ "role": "user", "content": "Hello from OmniKey!" }]
  }'
```

### Streaming

```typescript
const stream = await omni.chat.create({
  model: 'anthropic/claude-sonnet-4',
  messages: [{ role: 'user', content: 'Stream this.' }],
  stream: true,
});

for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0]?.delta?.content || '');
}
```

## Pricing

| Plan | Price | Includes |
|------|-------|----------|
| Starter | $10/mo per agent | 5 agents, 100K tokens/mo, 3 providers, basic analytics |
| Pro | $49/mo per agent | Unlimited agents, 1M tokens/mo, all providers, advanced analytics, health monitoring, 99.9% SLA |
| Enterprise | Custom | Dedicated infrastructure, BYOK, SSO, role-based access, custom SLAs, compliance packages |

## Key Features

- **Agent Health Monitoring**: Real-time dashboards tracking costs, latencies, tool call success rates, and runaway agent detection
- **Preset Configurations**: Save and reuse model configurations, temperature settings, and routing rules
- **API Catalogue**: Browse and connect to all available AI providers with BYOK (Bring Your Own Key) support
- **Platform Key Management**: Enterprise-grade key provisioning with automated rotation, rate limiting, anomaly detection, and compliance auditing
- **Internal Management Agents**: Autonomous worker bots (Key Sentinel, Cost Optimizer, Rotation Guardian, Rate Shield, Anomaly Detector, Compliance Auditor) that continuously monitor platform security and performance
- **Unified Billing**: Single invoice across all providers with cost attribution per agent, per model, per user

## Use Cases

- **AI Agent Developers**: Build agents that dynamically select the best model for each task without managing multiple API keys
- **Multi-Model Applications**: Applications that route requests to different providers based on cost, latency, or capability
- **Enterprise AI Platforms**: Centralized AI infrastructure management for organizations running agent fleets at scale
- **Rapid Prototyping**: Instantly test across providers without signup friction — one key, every model

## Links

- Website: https://omnikey.dev
- Documentation: https://omnikey.dev/docs
- Dashboard: https://omnikey.dev/dashboard
- Get Started: https://omnikey.dev/get-started

## Contact

For enterprise inquiries, partnerships, or support: Available through the OmniKey dashboard.

---

*This file follows the llms.txt standard (https://llmstxt.org) to help AI systems discover and understand OmniKey's capabilities.*