Workflow Concepts

Core concepts and architecture of HubFeed workflows

Workflow Concepts#

Understanding the core concepts behind HubFeed workflows will help you build more effective intelligence pipelines.

What is a Workflow?#

A workflow is a visual representation of a data processing pipeline. It defines:

  • Where data comes from (Sources)
  • How data is processed (Transforms)
  • Where data goes (Outputs)
  • Which credentials are used (Configuration)

Data Flow#

Data flows through your workflow in a directed graph:

Multi-Source Workflows#

You can aggregate data from multiple sources into a single pipeline:

Messages#

The fundamental unit of data in HubFeed is a Message. Each message contains:

  • id - Unique identifier
  • content - The main text content
  • title - Optional title
  • url - Source URL
  • author - Content author
  • created_at - When the content was created
  • metadata - Platform-specific data

Enrichments#

As messages pass through transforms, they accumulate enrichments:

json
{
  "enrichments": {
    "extracted_ips": ["192.168.1.1"],
    "extracted_emails": ["test@example.com"],
    "matched_keywords": ["security", "threat"]
  }
}

Workflow States#

Workflows can be in one of several states:

StateDescription
DraftNot yet activated
ActiveRunning and processing data
PausedTemporarily stopped
ErrorEncountered an error

Validation#

Before a workflow can be activated, it must pass validation:

  • All required fields must be configured
  • Connections must be valid
  • No circular dependencies

Important: Invalid workflows cannot be saved or activated.

Next Steps#