← Back to Docs

Quick Start Guide

Get from zero to 40-60% cheaper AI calls in under 2 minutes.

1

Create Your Account

Sign up for free — no credit card required. Beta has unlimited free usage.

2

Get Your API Key (30 seconds)

The fastest way — run one command:

# Python
pip install agentready-sdk && agentready init

# Node.js
npx agentready-sdk init

Or get your key at agentready.cloud/quick-key — it starts with ak_.

3

Two-Step: Compress Then Call Your LLM (Python)

Compress your messages with AgentReady, then call OpenAI directly with your own key — your OpenAI key never touches AgentReady:

import agentready
from openai import OpenAI
import os

# Step 1 — compress (only your AgentReady key)
result = agentready.compress(
    api_key=os.environ["AGENTREADY_API_KEY"],
    messages=[{"role": "user", "content": "Hello!"}]
)

# Step 2 — call OpenAI directly (your own key, never touches AgentReady)
client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])
response = client.chat.completions.create(
    model="gpt-4o",
    messages=result["messages"]
)
3b

Two-Step: Compress Then Call Your LLM (Node.js)

# Node.js
npm install agentready openai
import { compress } from 'agentready';
import OpenAI from 'openai';

// Step 1 — compress (only your AgentReady key)
const { messages, stats } = await compress({
  apiKey: process.env.AGENTREADY_API_KEY,
  messages: [{ role: 'user', content: 'Hello!' }],
});

// Step 2 — call OpenAI directly (your own key, never touches AgentReady)
const client = new OpenAI({ apiKey: process.env.OPENAI_API_KEY });
const res = await client.chat.completions.create({
  model: 'gpt-4o',
  messages,
});
3c

Or Monkey-Patch Existing Code

from agentready import patch_openai
import os
patch_openai(api_key=os.environ["AGENTREADY_API_KEY"])

# All existing OpenAI code now compresses automatically

🎉 That's it!

Compress your prompts with AgentReady, then call your LLM directly — saving 40-60% on tokens. Your provider key never touches AgentReady. Check your dashboard to see savings in real time.