dgi.intent_recognition

AI-powered intent recognition

AI-powered intent recognition for DGI conversations.

Uses AI to analyze user messages and determine their intent with better accuracy than keyword-based matching.

Inputs:

• message: The user's message (required)
• history: Previous messages in conversation (optional)
• available_workflows: List of workflow IDs to consider (optional)
• model: Optional model ID override (default: plugin.default_model)

Outputs (terminal state_data):

• recognized_intent: The identified intent (e.g., "deploy_site", "provision_ec2")
• intent_confidence: Confidence score (0.0-1.0)
• intent_params: Extracted parameters from the message
• suggested_workflow: Recommended workflow ID or null

Plugin required: context.get_plugin("ai")

Overview

Property	Value
Workflow type	Atomic
Library	App-dgi
Version	1.0

Input Schema

Field	Type	Required	Default	Description
message	string	Yes	—	User's message
history	list	No	—	Previous messages in conversation
available_workflows	list	No	—	List of workflow IDs to consider
model	string	No	—	Model ID override

Output Schema

Field	Type	Required	Default	Description
message	string	No	—	—
history	list	No	—	—
available_workflows	list	No	—	—
model	string	No	—	—
intent_confidence	float	No	—	—
intent_params	dict	No	—	—
intent_raw_response	dict	No	—	—
recognized_intent	string	No	—	—
suggested_workflow	string	No	—	—
failure_reason	string	No	—	—
failure_type	string	No	—	—
failed_action	string	No	—	—
failed_at_state	string	No	—	—
failed_step	string	No	—	—
failed_layer	string	No	—	—
error	string	No	—	—
error_type	string	No	—	—

States

State	Initial	Terminal	Success	Auto-advance	Description
pending	Yes	No	—	execute	—
completed	No	Yes	Yes	—	—
failed	No	Yes	No	—	—

State Diagram

Transitions

From	Action	To	Description
pending	execute	completed	—
* (any state)	fail	failed	—

API Usage

POST /api/workflows/start
Content-Type: application/json

{
  "workflow_type": "dgi.intent_recognition",
  "initial_data": {
    "message": "value"
  }
}