Skip to main content

​
Overview

The FlowX Knowledge Base integration enables you to ingest static documents and dynamic data feeds into contextual Knowledge Bases that AI agents can query when preparing responses. This integration provides a centralized repository for information that can be semantically searched and retrieved by AI agents during workflow execution.
Knowledge Bases use vector embeddings and semantic search to find the most relevant information for AI agent queries.

​
Key features

  • Centralized content management: Upload documents and ingest data from workflows into organized Knowledge Bases
  • Multiple content sources: Manage content by splitting it across different sources for better organization
  • Semantic search: AI agents can find relevant information using natural language queries
  • Testing capabilities: Test queries and operations in isolation before adding them to workflows
  • Content versioning: Append, replace, or delete content sources with full traceability
  • Relevance scoring: Understand which content chunks are most relevant for agent responses

​
How it works

​
High-level workflow

  1. Create Knowledge Base: Add a new Knowledge Base data source in Integration Designer
  2. Ingest content: Upload documents or ingest data from workflows into content sources
  3. Automatic chunking: Content is automatically split into chunks and indexed for semantic search
  4. Query in workflows: AI agents query the Knowledge Base to find relevant information
  5. Monitor usage: Track which chunks are used and their relevance scores in console logs

​
Main capabilities

​
Content ingestion

Manual upload

Upload PDF documents directly from the Knowledge Base admin interface

Workflow ingestion

Ingest JSON content dynamically from workflow variables

​
Content management

1

Organize by sources

Manage content by organizing it into separate content sources for better control and traceability
2

Append or replace

Update existing content sources by appending new information or replacing it entirely
3

Delete when needed

Remove content sources and all associated chunks when they’re no longer needed

​
AI agent integration

AI agents can use Knowledge Bases with the ReAct (Reasoning and Acting) model to:
  • Find relevant information based on user queries
  • Understand which content sources provided the information
  • See relevance scores for retrieved chunks
  • Make informed decisions based on contextual knowledge

​
Content sources

Content sources represent individual collections of information within a Knowledge Base. Each content source can be:
  • Manual: Files uploaded through the Knowledge Base admin interface
  • Workflow: Content ingested through workflow operations
Knowledge Base Content Sources

​
Content source lifecycle

Content sources progress through the following states:
  • Uploading: Content is being uploaded
  • Chunking: Content is being processed and split into chunks
  • Available: Content is ready and can be queried
  • Failed: Processing encountered an error
Chunks are small snippets of content that are indexed in the vector database for semantic search. When an AI agent queries a Knowledge Base, the most relevant chunks are returned based on:
  • Content similarity: Semantic meaning of the query vs. chunk content
  • Relevance score: Percentage indicating how relevant the chunk is (0-100%)
  • Metadata filters: Optional filters based on content source and metadata
You can test different query parameters to see which chunks are returned and their relevance scores before using the Knowledge Base in production workflows.

​
Use cases

​
Product documentation assistant

Create a Knowledge Base with product documentation and allow AI agents to answer customer questions based on the latest documentation.

​
Policy compliance

Upload company policies and compliance documents. AI agents can reference these when processing requests to ensure compliance.

​
Dynamic knowledge updates

Ingest data from external systems into Knowledge Bases through workflows, keeping the AI agent’s knowledge up-to-date automatically.

​
Multi-source information synthesis

Organize information across multiple content sources and let AI agents synthesize information from different sources to provide comprehensive answers.

​
Limitations

The current release has the following limitations:
Current Limitations:
  • Only PDF documents are supported for manual upload (images, PPT, DOC, XLS support coming in future releases)
  • Custom metadata cannot be defined (only system metadata is available)
  • AI agents can only use one Knowledge Base per Custom Agent node
  • Documents received from integrations cannot be directly ingested (Document Plugin links are treated as strings)
  • Content source renaming is not available

​
Creating a knowledge base

​
Prerequisites

Before you begin, ensure you have:
  • Access to FlowX Designer with appropriate permissions
  • A project where you want to add the Knowledge Base
  • Content ready to ingest (PDF documents or JSON data)

​
Setup steps

To add a new Knowledge Base data source:
1

Navigate to Integration Designer

Go to FlowX Designer → Workspaces → Your workspace → Projects → Your project → Integrations → Data Sources
2

Create new data source

Click Add New Data Source and select FlowX Knowledge Base as the data source type
3

Configure basic settings

Fill in the required fields with the following information
Add Knowledge Base

​
Configuration fields

​
Name

The unique identifier for your Knowledge Base within the project.
​
Name
string
required
Validation rules:
  • Required: Field cannot be empty
  • Uniqueness: Must be unique within the project
  • Special characters: Only letters, numbers, and the following characters are allowed: [], (), ., _, -
  • Length: Minimum 3 characters, maximum 50 characters
Validation messages:
  • “Name must be unique inside the project.”
  • “Name can only contain letters, numbers and the following special characters [] () . _ -”
  • “Name must contain at least 3 characters.”
  • “Name must contain maximum 50 characters.”
Choose a descriptive name that clearly indicates the purpose of the Knowledge Base, such as “Product Documentation” or “Customer Policies”.

​
Description

An optional description explaining the purpose and contents of the Knowledge Base.
​
Description
string
Provide additional context about what information this Knowledge Base contains and how it should be used.

​
Example configuration

Here’s an example of a well-configured Knowledge Base: 
Name: Product Documentation KB
Description: Contains all product documentation, user guides, and FAQ documents for the customer support AI agent

​
After creation

Once you’ve created a Knowledge Base, you’ll see three main sections:

Content Sources

View and manage all content sources ingested into the Knowledge Base

Chunks

Search and view the individual chunks created from your content

Operations

Test Knowledge Base operations before using them in workflows
Knowledge Base Dashboard

​
Common validation errors

Error: Name must be unique inside the project.
Solution: Choose a different name for your Knowledge Base. Each Knowledge Base must have a unique name within the project.
Error: Name can only contain letters, numbers and the following special characters [] () . _ -
Solution: Remove any special characters that are not in the allowed list. For example, use Customer_KB instead of Customer@KB.
Error: Name must contain at least 3 characters.
Error: Name must contain maximum 50 characters.
Solution: Ensure your Knowledge Base name is between 3 and 50 characters in length.

​
Best practices

Naming conventions:
  • Use descriptive names that indicate the content type (e.g., “HR Policies KB”, “Technical Documentation KB”)
  • Include the domain or department if managing multiple Knowledge Bases (e.g., “Sales Product Catalog”, “Support FAQ Database”)
  • Avoid generic names like “KB1” or “Test” in production environments
Organization:
  • Create separate Knowledge Bases for different domains or use cases
  • Group related content within the same Knowledge Base using content sources
  • Document the purpose of each Knowledge Base in the description field

​
Next steps

Managing Content

Upload documents and manage content sources

Using in Workflows

Ingest content and query Knowledge Bases from workflows

Testing Operations

Test queries and operations before production use

Integration Designer

Integration Designer and data sources

Custom Agent Nodes

Using AI agents in workflows

Integrations Overview

Integration ecosystem overview