Chatflows

This guide covers the management and usage of chatflows in the Flowise service.

Chatflows Overview

Chatflows are the core building blocks in Flowise that allow you to create conversational AI workflows using a visual builder. Each chatflow represents a complete conversation flow that can be embedded in your applications or accessed via API.

Chatflow Endpoints

List Chatflows

Retrieve a list of all available chatflows.

GET /api/v1/chatflows

curl "https://flowise.moodmnky.com/api/v1/chatflows" \
  -H "x-api-key: your_api_key"

Response:

{
  "chatflows": [
    {
      "id": "chatflow-1234",
      "name": "Customer Support Bot",
      "category": "Support",
      "deployed": true,
      "isPublic": true,
      "updatedAt": "2024-03-15T12:30:45Z"
    },
    {
      "id": "chatflow-5678",
      "name": "Product Recommender",
      "category": "Sales",
      "deployed": true,
      "isPublic": false,
      "updatedAt": "2024-03-20T09:15:22Z"
    }
  ]
}

Get Chatflow

Retrieve a specific chatflow by ID.

GET /api/v1/chatflows/{chatflowId}

curl "https://flowise.moodmnky.com/api/v1/chatflows/chatflow-1234" \
  -H "x-api-key: your_api_key"

Response:

{
  "id": "chatflow-1234",
  "name": "Customer Support Bot",
  "deployed": true,
  "isPublic": true,
  "category": "Support",
  "updatedAt": "2024-03-15T12:30:45Z",
  "nodes": [...],
  "edges": [...],
  "variables": [...]
}

Create Chatflow

Create a new chatflow.

POST /api/v1/chatflows

curl -X POST "https://flowise.moodmnky.com/api/v1/chatflows" \
  -H "x-api-key: your_api_key" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "New Support Bot",
    "category": "Support",
    "isPublic": true,
    "nodes": [...],
    "edges": [...]
  }'

Response:

{
  "id": "chatflow-9876",
  "name": "New Support Bot",
  "category": "Support",
  "deployed": false,
  "isPublic": true,
  "createdAt": "2024-04-01T10:25:12Z",
  "updatedAt": "2024-04-01T10:25:12Z"
}

Update Chatflow

Update an existing chatflow.

PUT /api/v1/chatflows/{chatflowId}

curl -X PUT "https://flowise.moodmnky.com/api/v1/chatflows/chatflow-9876" \
  -H "x-api-key: your_api_key" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Updated Support Bot",
    "category": "Support",
    "nodes": [...],
    "edges": [...]
  }'

Delete Chatflow

Delete a chatflow.

DELETE /api/v1/chatflows/{chatflowId}

curl -X DELETE "https://flowise.moodmnky.com/api/v1/chatflows/chatflow-9876" \
  -H "x-api-key: your_api_key"

Duplicate Chatflow

Duplicate an existing chatflow.

POST /api/v1/chatflows/{chatflowId}/duplicate

curl -X POST "https://flowise.moodmnky.com/api/v1/chatflows/chatflow-1234/duplicate" \
  -H "x-api-key: your_api_key" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Customer Support Bot (Copy)"
  }'

Response:

{
  "id": "chatflow-9999",
  "name": "Customer Support Bot (Copy)",
  "category": "Support",
  "deployed": false,
  "isPublic": true,
  "createdAt": "2024-04-01T10:25:12Z",
  "updatedAt": "2024-04-01T10:25:12Z"
}

Chat Prediction

Execute a chatflow prediction (non-streaming).

POST /api/v1/prediction/{chatflowId}

curl -X POST "https://flowise.moodmnky.com/api/v1/prediction/chatflow-1234" \
  -H "x-api-key: your_api_key" \
  -H "Content-Type: application/json" \
  -d '{
    "question": "Hello, how can you help me?",
    "overrideConfig": {},
    "sessionId": "session-123"
  }'

Stream Chat

Execute a chatflow prediction with streaming response.

POST /api/v1/prediction/{chatflowId}/stream

curl -X POST "https://flowise.moodmnky.com/api/v1/prediction/chatflow-1234/stream" \
  -H "x-api-key: your_api_key" \
  -H "Content-Type: application/json" \
  -d '{
    "question": "Tell me about your products",
    "overrideConfig": {},
    "sessionId": "session-123"
  }'

Stop Chat

Stop an active streaming chatflow execution.

POST /api/v1/prediction/{chatflowId}/stop

curl -X POST "https://flowise.moodmnky.com/api/v1/prediction/chatflow-1234/stop" \
  -H "x-api-key: your_api_key" \
  -H "Content-Type: application/json" \
  -d '{
    "sessionId": "session-123"
  }'

Deploy Chatflow

Deploy a chatflow to make it available for use.

POST /api/v1/chatflows/{chatflowId}/deploy

curl -X POST "https://flowise.moodmnky.com/api/v1/chatflows/chatflow-9876/deploy" \
  -H "x-api-key: your_api_key"

Response:

{
  "id": "chatflow-9876",
  "deployed": true,
  "updatedAt": "2024-04-01T10:30:45Z"
}

Undeploy Chatflow

Undeploy a chatflow to make it unavailable for use.

POST /api/v1/chatflows/{chatflowId}/undeploy

curl -X POST "https://flowise.moodmnky.com/api/v1/chatflows/chatflow-9876/undeploy" \
  -H "x-api-key: your_api_key"

Response:

{
  "id": "chatflow-9876",
  "deployed": false,
  "updatedAt": "2024-04-01T10:35:12Z"
}

Chatflow Parameters

Chatflow Object

Field	Type	Description
id	string	Unique identifier for the chatflow
name	string	Name of the chatflow
category	string	Category for organization purposes
deployed	boolean	Whether the chatflow is deployed and available for use
isPublic	boolean	Whether the chatflow is accessible without authentication
nodes	array	Array of node objects that make up the chatflow
edges	array	Array of edge objects connecting the nodes
variables	array	Array of variable objects for the chatflow
createdAt	string	Creation timestamp
updatedAt	string	Last update timestamp

Node Object

Field	Type	Description
id	string	Unique identifier for the node
type	string	Type of the node (e.g., “textInput”, “llmChain”)
position	object	X and Y coordinates of the node in the builder
data	object	Configuration data for the node

Edge Object

Field	Type	Description
id	string	Unique identifier for the edge
source	string	ID of the source node
target	string	ID of the target node
sourceHandle	string	Output handle of the source node
targetHandle	string	Input handle of the target node

Usage Examples

List and Filter Chatflows

// List all chatflows
const response = await client.flowise.listChatflows();
console.log(`Found ${response.chatflows.length} chatflows`);

// Filter by category
const supportChatflows = response.chatflows.filter(
  flow => flow.category === "Support"
);
console.log(`Found ${supportChatflows.length} support chatflows`);

// Get only deployed chatflows
const deployedChatflows = response.chatflows.filter(
  flow => flow.deployed === true
);
console.log(`Found ${deployedChatflows.length} deployed chatflows`);

Manage Chatflow Deployment Status

// Deploy a chatflow
async function deployChatflow(chatflowId) {
  try {
    const result = await client.flowise.deployChatflow(chatflowId);
    console.log(`Chatflow ${chatflowId} deployed successfully`);
    return result;
  } catch (error) {
    console.error(`Failed to deploy chatflow: ${error.message}`);
    throw error;
  }
}

// Undeploy a chatflow
async function undeployChatflow(chatflowId) {
  try {
    const result = await client.flowise.undeployChatflow(chatflowId);
    console.log(`Chatflow ${chatflowId} undeployed successfully`);
    return result;
  } catch (error) {
    console.error(`Failed to undeploy chatflow: ${error.message}`);
    throw error;
  }
}

// Toggle deployment status
async function toggleChatflowDeployment(chatflowId) {
  const chatflow = await client.flowise.getChatflow(chatflowId);
  
  if (chatflow.deployed) {
    return await undeployChatflow(chatflowId);
  } else {
    return await deployChatflow(chatflowId);
  }
}

Create and Update Chatflows

// Create a simple chatflow with text input and LLM response
async function createSimpleChatbot(name, category, apiKey) {
  const chatflowData = {
    name,
    category,
    isPublic: true,
    nodes: [
      {
        id: "textInput",
        type: "textInput",
        position: { x: 100, y: 200 },
        data: { text: "Hello! How can I help you today?" }
      },
      {
        id: "llmChain",
        type: "llmChain",
        position: { x: 400, y: 200 },
        data: {
          model: "gpt-3.5-turbo",
          apiKey: apiKey,
          temperature: 0.7
        }
      }
    ],
    edges: [
      {
        id: "edge-1",
        source: "textInput",
        target: "llmChain",
        sourceHandle: "output",
        targetHandle: "input"
      }
    ]
  };
  
  return await client.flowise.createChatflow(chatflowData);
}

// Update chatflow with a new node
async function addMemoryToExistingChatflow(chatflowId) {
  // Get existing chatflow
  const chatflow = await client.flowise.getChatflow(chatflowId);
  
  // Add memory node
  const memoryNode = {
    id: "memoryNode",
    type: "bufferMemory",
    position: { x: 250, y: 300 },
    data: { memoryKey: "chat_history" }
  };
  
  // Add new edges
  const newEdges = [
    {
      id: "edge-to-memory",
      source: "textInput",
      target: "memoryNode",
      sourceHandle: "output",
      targetHandle: "input"
    },
    {
      id: "edge-from-memory",
      source: "memoryNode",
      target: "llmChain",
      sourceHandle: "output",
      targetHandle: "memory"
    }
  ];
  
  // Update chatflow
  chatflow.nodes.push(memoryNode);
  chatflow.edges = [...chatflow.edges, ...newEdges];
  
  return await client.flowise.updateChatflow(chatflowId, chatflow);
}

Best Practices

Chatflow Design
- Keep chatflows focused on specific use cases
- Use clear naming conventions
- Organize with meaningful categories
- Document the purpose of each chatflow
Deployment Management
- Test thoroughly before deploying
- Use deployment toggles for maintenance
- Monitor usage after deployment
- Implement versioning for important changes
Security Considerations
- Use isPublic flag judiciously
- Implement proper authentication
- Validate and sanitize user inputs
- Review node configurations for sensitive data
Performance Optimization
- Minimize complex node arrangements
- Use caching where appropriate
- Monitor response times
- Implement rate limiting for public chatflows

Advanced Features

Chatflow Variables

You can define variables in your chatflows that can be accessed and modified across nodes.

// Update chatflow to include variables
const chatflowWithVariables = {
  // ... existing chatflow data
  variables: [
    {
      name: "userName",
      type: "string",
      defaultValue: "Guest"
    },
    {
      name: "userPreferences",
      type: "json",
      defaultValue: { language: "en", theme: "light" }
    }
  ]
};

await client.flowise.updateChatflow(chatflowId, chatflowWithVariables);

Chatflow Export and Import

// Export a chatflow
async function exportChatflow(chatflowId) {
  const response = await client.flowise.exportChatflow(chatflowId);
  const json = JSON.stringify(response);
  // Save to file or database
  return json;
}

// Import a chatflow
async function importChatflow(chatflowJson) {
  const chatflowData = JSON.parse(chatflowJson);
  // Modify as needed
  chatflowData.name = `${chatflowData.name} (Imported)`;
  return await client.flowise.createChatflow(chatflowData);
}

​Chatflows

​Chatflows Overview

​Chatflow Endpoints

​List Chatflows

​Get Chatflow

​Create Chatflow

​Update Chatflow

​Delete Chatflow

​Duplicate Chatflow

​Chat Prediction

​Stream Chat

​Stop Chat

​Deploy Chatflow

​Undeploy Chatflow

​Chatflow Parameters

​Chatflow Object

​Node Object

​Edge Object

​Usage Examples

​List and Filter Chatflows

​Manage Chatflow Deployment Status

​Create and Update Chatflows

​Best Practices

​Advanced Features

​Chatflow Variables

​Chatflow Export and Import

​Support & Resources

Chatflows

Chatflows Overview

Chatflow Endpoints

List Chatflows

Get Chatflow

Create Chatflow

Update Chatflow

Delete Chatflow

Duplicate Chatflow

Chat Prediction

Stream Chat

Stop Chat

Deploy Chatflow

Undeploy Chatflow

Chatflow Parameters

Chatflow Object

Node Object

Edge Object

Usage Examples

List and Filter Chatflows

Manage Chatflow Deployment Status

Create and Update Chatflows

Best Practices

Advanced Features

Chatflow Variables

Chatflow Export and Import

Support & Resources