Customer support

1 Workflows1 Response Formats1 Backend StepsSession Memory

Session Memory Chatbot

Runs a real-time chatbot that remembers earlier messages in the same conversation and returns structured, UI-ready replies.

Works with

Primary
Backup

Pick your providers and models. The download is customized to your selections.

Best for

  • Customer-facing chatbots
  • Product and onboarding assistants
  • Support teams
Download JSONSee the integration guide
Source code on GitHub

How the data flows

Requests move between your backend and the AI workflows automatically. Here's the exact path a request takes.

Customer request arrives

AI workflow

mr_chatbot_workflow

Produces structured real-time replies while remembering the current conversation

Backend event

webhook_received

Customer receives the response

What you get

Every template is built from two kinds of pieces: Workflows and Response Formats.

Workflow

mr_chatbot_workflow

Produces structured real-time replies while remembering the current conversation

Response format

chatbot_response

A structured chatbot reply with conversation insights and follow-up actions

replysummarysentimentconfidencetopicsaction_items

Details

The full specification of each workflow, with plain-English explanations. Everything below is set in the downloaded file and can be changed after import.

mr_chatbot_workflow

Produces structured real-time replies while remembering the current conversation

Model & backupsLearn more
openai · gpt-5.6-lunaBackup 1: anthropic · claude-haiku-4-5-20251001

The primary model plus automatic fallback models if the primary is unavailable.

Response formatLearn more
chatbot_responseReturns a structured object instead of free-form text.

The structured JSON this workflow returns. Fields are defined by you and enforced by ModelRiver.

Customer data it usesLearn more
None - no customer data required

Fields you include in each request so the workflow can personalise its answer.

Session memoryLearn more
Enabled

Remembers earlier turns when your app sends the returned session_id. Request-body logging must be enabled for the project.

Automated pipelineLearn more
webhook_received

Steps that run automatically around this workflow, in order.

ModeLearn more
Production mode

Runs live against your connected AI provider.

How to use it

  1. 1

    Download the template

    The file is a portable JSON bundle - safe to share and easy to read.

  2. 2

    Open a project in ModelRiver

    Sign in, pick a project, and make sure the AI providers used by this template are connected. You can connect them from the Providers section.

  3. 3

    Import it

    Open the Import button in your project, paste or upload the file, review the preview, and confirm. Everything is created atomically, so nothing is overwritten. How import & export works

Template file

Technical users can inspect the raw bundle below.

View full template JSON
modelriver-session-memory-chatbot-template-v1.json3.2 KB
{
  "version": 1,
  "structures": [
    {
      "name": "chatbot_response",
      "description": "A structured chatbot reply with conversation insights and follow-up actions",
      "example": {
        "reply": "Your name is Vishal, you own Sunrise Bakes, and your business is in Bangalore.",
        "summary": "Recalled the user's name, business, and city from this conversation.",
        "sentiment": "neutral",
        "confidence": "high",
        "topics": [
          "personal profile",
          "business details"
        ],
        "action_items": []
      },
      "schema": {
        "type": "object",
        "properties": {
          "reply": {
            "type": "string"
          },
          "summary": {
            "type": "string"
          },
          "sentiment": {
            "type": "string",
            "enum": [
              "positive",
              "neutral",
              "negative",
              "mixed"
            ]
          },
          "confidence": {
            "type": "string",
            "enum": [
              "high",
              "medium",
              "low"
            ]
          },
          "topics": {
            "type": "array",
            "items": {
              "type": "string"
            }
          },
          "action_items": {
            "type": "array",
            "items": {
              "type": "object",
              "properties": {
                "task": {
                  "type": "string"
                },
                "priority": {
                  "type": "string",
                  "enum": [
                    "high",
                    "medium",
                    "low"
                  ]
                }
              },
              "required": [
                "task",
                "priority"
              ],
              "additionalProperties": false
            }
          }
        },
        "required": [
          "reply",
          "summary",
          "sentiment",
          "confidence",
          "topics",
          "action_items"
        ],
        "additionalProperties": false
      }
    }
  ],
  "workflows": [
    {
      "name": "mr_chatbot_workflow",
      "description": "Produces structured real-time replies while remembering the current conversation",
      "provider": "openai",
      "model": "gpt-5.6-luna",
      "backups": [
        {
          "provider": "anthropic",
          "model": "claude-haiku-4-5-20251001"
        }
      ],
      "request_type": "chat",
      "system_instructions": "Act as a helpful, concise customer-facing assistant. Use relevant facts from earlier turns in the current session when answering. If a fact is not present in the conversation, say that you do not know instead of inventing it. Do not claim to remember information outside the supplied session. Return a direct reply plus a short summary, sentiment, confidence, topics, and any clear follow-up action items.",
      "structure": "chatbot_response",
      "customer_fields": [],
      "test_mode": false,
      "sessions_enabled": true,
      "backend_pipeline": {
        "enabled": true,
        "events": [
          {
            "name": "webhook_received"
          }
        ]
      }
    }
  ]
}

Looking for something else?

More templates are on the way. In the meantime, you can build your own workflows from scratch in the ModelRiver console.

Browse all templatesRead the workflow guide