← All examples

Conversation Agent

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

agentconversationstreaming
Deploy this example

Build an agent that supports multi-turn conversations with managed message history. Ideal for chatbots, customer support agents, and interactive assistants.

What's included

  • Conversation type — implements the predefined conversation input/output schemas
  • Message history — Reminix manages session state and message history automatically
  • Streaming responses — real-time token streaming via SSE
  • Python setup — fully typed with Pydantic models and the Reminix SDK

When to use this

Pick the Conversation Agent example when you want to:

  • Build a multi-turn conversational bot
  • Create a customer support assistant
  • Develop an interactive agent that maintains context across messages

Project structure

├── agents/
│   └── conversation.py   # Conversation handler with history
├── main.py               # Entry point with serve()
├── pyproject.toml
└── reminix.config.toml   # Reminix configuration

Quick start

After deploying, start a conversation using the Reminix API:

curl -X POST https://api.reminix.com/v1/agents/support-bot/chat \
  -H "Authorization: Bearer $REMINIX_API_KEY" \
  -d '{"message": "Hi there! What can you help me with?"}'