InsightFlow Introduction

1. What is InsightFlow?

In GoInsight.ai, InsightFlow is a no-code solution that allows enterprise users to construct AI workflow automation. It enables users to visually orchestrate various nodes into a seamless and complete process from data collection to execution.

By utilizing InsightFlow, enterprises can efficiently integrate AI capabilities and business logic, automating the entire process to streamline operations and increase efficiency.

Key concept

• Nodes: Nodes are the key components in the workflow, each node is responsible for performing a specific task.
• Variables: Variables are used in workflow to store and transfer data, allowing workflow to share information between different nodes.

Quick Guide:

2. Nodes

There are many built-in nodes in GoInsight.AI platform.

Click and learn more:

• 
  Start Node
  Determine the entry of the workflow.
• 
  End Node
  Determine the exit of the workflow.
• 
  LLM
  Leverage large language models to process complex natural language output.
• 
  Knowledge Base Retrieval
  Retrieve relevant information from the internal knowledge base.
• 
  Answer
  Set a reply in a chat conversation.
• 
  Agent
  Autonomous, able to perceive, reason and utilize tools to accomplish tasks.
• 
  Document Read
  Read the document content in the knowledge base.
• 
  Document Write
  Write information into a knowledge base document.
• 
  HTTP Request
  Interact with external APIs/services by sending HTTP requests.
• 
  KnowledgeFocus LLM
  A specialized LLM based on knowledge-base-specific Q&A.
• 
  IF/ELSE
  Branches the process based on conditions such as IF, ELSE, etc.
• 
  Question Classifier
  Use LLMs to perform semantic analysis and categorize text inputs.
• 
  Branch Aggregator
  Combines outputs from different branches.
• 
  Multi-branch Selector
  Select the corresponding branch to execute subsequent logical operations
• 
  Iteration
  Executes repeatedly and combines all outputs into a single list.
• 
  Auto-Continue
  Automatically proceed to the next round of chat conversation.
• 
  Template
  A tool for creating customized text.
• 
  Code
  Run Python code within a workflow
• 
  JSON Variable Extractor
  Extract specified fields (values) from input data in JSON format
• 
  Variable Assigner
  Set specific values to variables.
• 
  Variable Extractor
  Analyze and interpret natural language input.
• 
  Tool Call
  Reuse an existing Service Flow as a Tool in other workflow.

3. Variables

Variables can be divided into three different types based on their scope and purpose.

3.1 System Variables

System variables are predefined parameters at the system level and can be accessed by other nodes.

Interactive Flow provides the following system variables:

Service Flow provides the following system variables:

3.2 Conversation Variable

Conversation variables are variables used to store and manage data throughout a conversation or session. They help maintain context, track user inputs, and manage the flow of interaction.

3.3 Internal Variable

Internal Variable that are used internally within a system or application. They are not exposed to the user and are typically used to manage internal processes or sensitive information.