Skip to main content
1

Install the SDK

Install the Reminix client SDK for your language.
npm install @reminix/sdk
2

Get your API key

Go to the Reminix Dashboard, create a project, and copy your API key from the project settings page.
Your API key starts with reminix_sk_ and is scoped to a single project. Keep it secret and never expose it in client-side code.
Every project includes built-in LLM access (OpenAI, Anthropic, Google) — enough to experiment and run examples. No provider keys needed to get started. You can add your own keys later in project settings.
3

Deploy your first agent

A new project is empty — you need at least one agent before you can invoke anything. Pick whichever path is fastest:
  • Start from a template (recommended) — pick one from reminix.com/templates and click Deploy. You get a working agent in your project in about thirty seconds, with the source forked into your own GitHub repo so you can customize it.
  • Connect your own GitHub repo — see Deploy from GitHub. Push code, get a live agent in a couple of minutes.
  • Use the CLI — write a server.ts or server.py with one agent() (TS) or @agent (Python) definition and a serve() call, then run reminix deploy from the project directory.
See Creating Agents in TypeScript or Creating Agents in Python for the smallest possible handler.The rest of this page assumes your agent is named my-agent. Substitute your agent’s actual name in the examples below.
4

Invoke an agent

Send a one-shot request to your deployed agent.
import Reminix from "@reminix/sdk"

const client = new Reminix({ apiKey: "your-api-key" })

const response = await client.agents.invoke("my-agent", {
  input: { prompt: "Summarize this document" },
})

console.log(response.output)
5

Chat with an agent

Start a multi-turn conversation. Use conversation_id to continue an existing conversation.
const chat = await client.agents.chat("my-agent", {
  messages: [{ role: "user", content: "What's the weather in SF?" }],
})

console.log(chat.output)

// Continue the conversation
const followUp = await client.agents.chat("my-agent", {
  messages: [{ role: "user", content: "What about tomorrow?" }],
  conversation_id: chat.conversation_id,
})
6

Stream responses

Pass stream: true to get real-time Server-Sent Events as the agent generates output.
const stream = await client.agents.invoke("my-agent", {
  input: { prompt: "Write a poem" },
  stream: true,
})

for await (const event of stream) {
  if (event.type === "text_delta") {
    process.stdout.write(event.delta)
  }
}

​
Next steps

Agents

Understand agent types, handlers, and metadata.

TypeScript: Creating Agents

Build and deploy agents with the TypeScript runtime.

Python: Creating Agents

Build and deploy agents with the Python runtime.