Mandala
Triggers

Starter

Manually initiate workflow execution with input parameters

The Starter block allows manual workflow execution with two input modes: structured parameters or conversational chat.

Execution Modes

Choose your input method from the dropdown:

  • Structured inputs: Define specific parameters (text, number, boolean, JSON, file, date)
  • Form interface: Users fill out a form with predefined fields
  • API friendly: Perfect for programmatic execution

Configure input parameters that will be available when deploying as an API endpoint.

  • Natural language: Users type questions or requests
  • start.input variable: Captures all user input as <start.input>
  • start.conversationId: Access conversation ID as <start.conversationId>
  • Conversational: Ideal for AI-powered workflows

Chat with your workflow and access input text, conversation ID, and uploaded files for context-aware responses.

Using Chat Variables

In Chat mode, access user input and conversation context through special variables:

# Reference the chat input, conversation ID, and files in your workflow
user_message: "<start.input>"
conversation_id: "<start.conversationId>"
uploaded_files: "<start.files>"
  • <start.input> - Contains the user's message text
  • <start.conversationId> - Unique identifier for the conversation thread
  • <start.files> - Array of files uploaded by the user (if any)

API Execution

curl -X POST "https://api.ayantram.com/dev/v1/workflows/{id}/start" \
  -H "Authorization: Bearer {api-key}" \
  -d '{"parameters": {"userId": "123", "action": "process"}}'
curl -X POST "https://api.ayantram.com/dev/v1/workflows/{id}/start" \
  -H "Authorization: Bearer {api-key}" \
  -d '{"input": "Analyze Q4 sales data"}'

Starter blocks are ideal for testing workflows and user-initiated tasks. For automated execution, use Schedule or Webhook triggers.

Starter