Your agent works.
The API shouldn't take two weeks.

Auth, streaming, validation, rate limiting, CORS, monitoring — that's weeks of infra work. Or one serve() call with Reminix.

Define it. Deploy it.

A schema and a handler. Use whatever LLM or framework you want — Reminix never touches your agent's internals.

  • Production REST API with auth and validation
  • SSE streaming with backpressure
  • Schema validation on input and output
  • Request tracing, error tracking, latency metrics
agents/support-bot.ts
import OpenAI from "openai"
import { agent } from "@reminix/runtime"

const openai = new OpenAI()

export const supportBot = agent("support-bot", {
  type: "chat",
  description: "Customer support assistant",
  handler: async (input) => {
    const response = await openai.chat
      .completions.create({
        model: "gpt-4o",
        messages: input.messages,
      })
    return response.choices[0].message.content
  },
})

One deploy. Four endpoints.

Reminix generates production endpoints from your agent definition.

POST/v1/agents/{name}/invoke
POST/v1/agents/{name}/chat
POST/v1/agents/{name}/workflow
GET/v1/agents/{name}

Or call it from your terminal.

Invoke agents, stream responses, and run multi-turn conversations directly from the CLI. Pipe output into scripts, CI/CD pipelines, or other AI agents.

Terminal
# Invoke an agent
$reminix agent invoke support-bot \
--prompt "How do I reset my password?"
# Multi-turn chat
$reminix agent chat support-bot \
--message "Hello" --stream

Simple tasks. Conversations. Workflows.

Not every agent is a chatbot. Reminix supports all three patterns out of the box.

Invoke

Send input, get output. Supports streaming.

Data processing, extraction, analysis

Conversations

Multi-turn with managed sessions.

Support bots, assistants, research

Workflows

Stateful processes that pause and resume.

Pipelines, approvals, orchestration

Connect to 20+ services.

Google Calendar, Slack, GitHub, Notion — Reminix handles the full OAuth flow. Your code just gets a valid token.

using-connections.ts
// Get a fresh token — auto-refreshed
const { access_token } = await client
  .oauthConnections.getToken("google")

// Use Google's own SDK directly
const calendar = google.calendar({
  version: "v3",
  auth: access_token,
})

const events = await calendar.events.list({
  calendarId: "primary",
  timeMin: new Date().toISOString(),
})

Bring your own framework.

LangChain, Vercel AI SDK, OpenAI, Anthropic, Gemini, or custom code. Wrap it in a Reminix agent and deploy.

LangChainLangChain
Vercel AI SDKVercel AI SDK
OpenAIOpenAI
AnthropicAnthropic
GeminiGemini

Agents for the problems you actually solve.

Ready-to-use agent templates for the most common use cases. Fork, deploy, and call the REST API from any app.

Conversation Agent

Agent implementing the conversation type for multi-turn sessions with message history and streaming.

TSPY

Research Assistant

Conversation agent that uses tools to search, summarize, and synthesize information across multiple sources.

TSPY

Customer Support Agent

Conversation agent designed for customer support. Multi-turn chat with context awareness and escalation logic.

TSPY

Meeting Scheduler

Full project with a conversation agent, Google Calendar tool, and Gmail tool. The agent schedules meetings, checks availability, and sends invitations.

TSPY

Sales Assistant

Full project with a conversation agent, HubSpot tool, and Gmail tool. The agent manages leads, updates CRM records, and sends follow-up emails.

TSPY

Workflow Agent

Agent implementing the workflow type for multi-step orchestration with pause and resume.

TSPY
Browse all agent templates

Your agent deserves a real API.

Deploy in five minutes. Start free — no credit card required.