What is Reactive Agents

From Prototype to Reliable AI, Quickly

Powered by the Agent and Skill concepts and supporting over 35+ AI providers, Reactive Agents delivers complete provider freedom, intelligent routing, and best-in-class automatic optimization.

What Makes Reactive Agents Unique

AI Agent Optimization with Multi-Provider Support

Reactive Agents is built around Agents and Skills — proven concepts for managing AI systems. We use familiar terminology and workflows, but with the key advantage of supporting multiple AI providers:

Our Key Advantages

Multi-Provider Support: Use OpenAI, Google Gemini, xAI Grok, and 34 other AI providers
Continuous Optimization & Learning: Continuously improves performance through intelligent analysis and feedback loops.
Smart Configuration Selection: Automatically identifies the best configuration for your LLM
Unified API: A unified API that works seamlessly with any of our supported AI provider
Advanced Analytics: Comprehensive evaluation and performance tracking

For detailed information about performance optimization and configuration parameters, see our Performance guide.

Product Overview

🎥

Product Demo Video

Coming Soon

Integration Example

Get started with Reactive Agents in minutes - here's all you need to start routing your AI requests:

import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: 'reactive-agents',
  baseURL: 'http://localhost:3000/v1',
});

const response = await client
  .withOptions({
    defaultHeaders: {
      'ra-config': JSON.stringify({
        targets: [{ optimization: 'auto' }],
        agent_name: 'My Agent',
        skill_name: 'General',
      }),
    },
  })
  .chat.completions.create({
    model: 'gpt-5',
    messages: [{ role: 'user', content: 'Hello!' }],
  });

const response = await fetch('http://localhost:3000/v1/chat/completions', {
  method: 'POST',
  headers: {
    'Authorization': `Bearer ${process.env.BEARER_TOKEN ?? 'reactive-agents'}`,
    'Content-Type': 'application/json',
    'ra-config': JSON.stringify({
      targets: [{ optimization: 'auto' }],
      agent_name: 'My Agent',
      skill_name: 'General',
    }),
  },
  body: JSON.stringify({
    model: 'gpt-5',
    messages: [{ role: 'user', content: 'Hello!' }],
  }),
});

const data = await response.json();

curl -X POST http://localhost:3000/v1/chat/completions \
  -H "Authorization: Bearer ${process.env.BEARER_TOKEN:-reactive-agents}" \
  -H "Content-Type: application/json" \
  -H "ra-config: {\"targets\":[{\"optimization\":\"auto\"}" \
  -d '{
    "model": "gpt-5",
    "messages": [
      {"role": "user", "content": "Hello!"}
    ]
  }'

Ready to get started?

Use Docker Compose

Run Local Instance

Frequently Asked Questions

Will Reactive Agents increase the latency of my API requests?

Reactive Agents adds minimal latency (typically 20-40ms) compared to direct API calls. This slight increase is often offset by the benefits of monitoring, evaluation, and optimization capabilities.

Is my data secure?

Reactive Agents is designed with security in mind. All data is encrypted in transit and at rest. You can run Reactive Agents locally for complete data control, or use our managed hosting options for enterprise deployments.

Does Reactive Agents support all AI providers?

Reactive Agents supports 35+ AI providers in total, with 4 key providers (OpenAI, Google Gemini, xAI Grok, and Anthropic) having comprehensive examples. The remaining 33 providers have infrastructure support and can be configured through the API.

How does Reactive Agents help improve my AI systems?

Reactive Agents provides continuous monitoring, evaluation tools, and performance analytics to help you identify issues and optimize your AI applications over time. It automatically tracks agent performance and provides insights for improvement.

What's the difference between Reactive Agents and direct API calls?

Reactive Agents adds a layer of intelligence on top of your AI providers, offering monitoring, evaluation, agent management, and optimization capabilities that aren't available with direct API calls.

How is Reactive Agents different from single-provider agent platforms?

Reactive Agents uses familiar agent and skill concepts - same terminology, same workflows, same approach. The key difference is multi-provider support. You get all the benefits of agent management platforms plus the ability to use 35+ different AI providers, automatic optimization, and advanced analytics across all providers.

How does Reactive Agents's optimization work?

Reactive Agents automatically groups similar conversations and identifies the best configuration for your LLM. The system learns from evaluation feedback and continuously improves performance without manual tuning. This creates a self-improving AI system that adapts to your specific use cases.