Workflow API - Quick Integration Guide

Workflows allow you to orchestrate multiple agents, and these workflows can be called externally via APIs at any time. In this guide, we will walk you through the steps.

Quick steps to execute Workflows externally

Step 1: Design Your Workflow in Agent Studio UI

What You’re Doing: Creating your workflow visually instead of writing complex JSON manually. Why This Matters: The Agent Studio visual builder automatically generates the correct JSON structure for you - no need to understand complex workflow syntax or risk JSON errors. How to Do It:

Open Agent Studio visual workflow builder
Drag and drop nodes (agents, APIs, conditionals) onto the canvas
Connect nodes with lines to define the flow
Configure each node’s parameters in the UI
CRITICAL: Click on Workflow API
Copy the entire workflow_data JSON structure

Step 2: Execute Your Workflow via run-dag API

What You’re Doing: Taking the JSON from Step 1 and executing it with real data. Why This Matters: This is where your visual workflow becomes a running process that does actual work (processes data, calls APIs, makes decisions). How to Do It:

curl -X POST "https://lao.studio.lyzr.ai/run-dag/" \
  -H "Content-Type: application/json" \
  -d '{
    "workflow_data": {
      // PASTE THE COMPLETE JSON FROM STEP 1 HERE
      // Everything from "Export JSON" goes here
      "tasks": [
        {
          "name": "process_data",
          "function": "agent", 
          "params": {
            "config": {
              "agent_id": "your_agent_id",
              "api_key": "your_agent_key"
            },
            "query": "Process this customer inquiry"
          }
        }
      ],
      "flow_name": "Quick_Test",
      "run_name": "test_001"
    },
    "inputs": {
      // These are the runtime values - what your workflow processes
      "user_input": "Hello world",
      "customer_email": "user@company.com"
    }
  }'

What Happens: The workflow engine executes each node in the correct order, passing data between them, and returns the final results. Key Points:

workflow_data: The complete JSON from your visual builder (Step 1)
inputs: The actual data you want to process (overrides default values)
Response contains the results from each node in your workflow

Step 3: Monitor Execution with WebSocket Events

What You’re Doing: Getting real-time updates as your workflow runs. Why This Matters: For enterprise use, you need to know immediately when workflows complete, fail, or need attention. Don’t wait for polling - get instant notifications. How to Do It:

// Connect to the specific workflow execution
const ws = new WebSocket('wss://lao-socket.studio.lyzr.ai/ws/Quick_Test/test_001');

ws.onopen = () => {
  console.log('✅ Connected - monitoring workflow execution');
};

ws.onmessage = (event) => {
  const data = JSON.parse(event.data);
  console.log(`${data.event_type}: ${data.task_name}`);
  
  // Handle different event types
  switch(data.event_type) {
    case 'task_completed':
      console.log('Node finished:', data.output);
      break;
    case 'flow_completed':
      console.log('🎉 Workflow done! Results:', data.output);
      // Process final results, update UI, notify users
      break;
    case 'task_failed':
      console.error('❌ Node failed:', data.error);
      // Handle errors, retry, alert ops team
      break;
  }
};

ws.onerror = (error) => {
  console.error('WebSocket error:', error);
};

What You Get:

Real-time progress updates as each node executes
Immediate error notifications if something fails
Final results as soon as the workflow completes
Ability to update users and trigger other systems instantly

Executing workflow via run-dag Endpoint

What This Is: The core API endpoint that takes your visual workflow and executes it with real data. URL: https://lao.studio.lyzr.ai/run-dag/
Method: POST
Purpose: Execute any workflow with real-time monitoring Why You Use This: This is where your designed workflow becomes a running automation that processes data, calls APIs, makes decisions, and produces results.

Request Structure Explained

{
  "workflow_data": {
    // This entire block comes from Agent Studio's "Export JSON"
    "tasks": [...],           // Your workflow nodes (agents, APIs, etc.)
    "flow_name": "string",    // Workflow name from Studio
    "run_name": "string",     // Unique execution ID (you choose this)
    "default_inputs": {...},  // Default values from input nodes
    "edges": [...]            // How nodes connect (auto-generated)
  },
  "inputs": {
    // The actual data you want to process RIGHT NOW
    "customer_message": "Help with billing",
    "priority": "high"
  }
}

Key Concept:

workflow_data = Your workflow template (same every time)
inputs = The specific data for THIS execution (changes each time)

How to Get workflow_data JSON

🎯 Method 1: Agent Studio Visual Builder ⭐ Always Use This What You’re Doing: Getting the perfect JSON without any manual work. Step-by-Step:

Open Agent Studio workflow builder
Design your workflow visually (drag, drop, connect nodes)
Click “Export JSON” or “Get API Code” button in the UI
Copy the ENTIRE workflow_data JSON structure
Paste it into your API calls

Why This Works: The visual builder generates syntactically perfect, validated JSON that includes all the complex node configurations, dependencies, and connections.

🔧 Method 2: Workflow Management API (For Existing Workflows) What You’re Doing: Retrieving a workflow you already created and saved.

# Get existing workflow JSON
curl -X GET "{BASE_URL}/v3/workflows/{flow_id}" \
  -H "x-api-key: YOUR_API_KEY"

# The response contains a 'flow_data' field
# Use that flow_data as your workflow_data in run-dag calls

When You Use This: When you have workflows already saved in the system and want to execute them programmatically.

❌ Method 3: Manual Construction (NEVER Do This) What This Means: Writing the complex JSON structure by hand. Why You Don’t Do This:

Extremely complex JSON structure with nested dependencies
Easy to make syntax errors that break execution
Hard to maintain and debug
The visual builder does this perfectly for you

Exception: Only for very simple, single-node workflows for testing.

Response Structure

Initial Response (Immediate):

{
  "status": "processing",
  "task_id": "0d3c400a-d35b-43db-9e9f-655d857d8594"
}

What This Means:

Your workflow has been accepted and is now running
Use the task_id to track the execution
Connect to WebSocket using your flow_name and run_name to monitor progress
The workflow continues executing asynchronously

Final Results (via WebSocket):

{
  "event_type": "flow_completed",
  "status": "completed",
  "results": {
    "node_name": { "output": "..." }
  },
  "execution_time": 12.34,
  "task_id": "0d3c400a-d35b-43db-9e9f-655d857d8594"
}

Getting Final Results via Task Status API

What This Is: An alternative way to get workflow results after execution completes, using the task_id from the initial response. When to Use This:

When you can’t use WebSocket connections (firewall restrictions, etc.)
When you need to retrieve results later (hours/days after execution)
As a backup method to ensure you capture results

Task Status Endpoint

URL: https://lao.studio.lyzr.ai/task-status/{task_id}
Method: GET
Purpose: Retrieve the final results of a completed workflow

Example Request

curl -X 'GET' \
  'https://lao.studio.lyzr.ai/task-status/0d3c400a-d35b-43db-9e9f-655d857d8594' \
  -H 'accept: application/json'

Response While Processing

Important: This endpoint returns processing status while the workflow is still running. You’ll need to poll periodically or use WebSocket for real-time updates.

{
  "status": "processing",
  "task_id": "0d3c400a-d35b-43db-9e9f-655d857d8594"
}

Response After Completion

Once the workflow finishes, you get the complete results from all nodes:

{
  "status": "completed",
  "results": {
    "agent_yellowidea": "Theme Prompt: Imagine a clandestine society of cats who, by moonlight and under carpets, have taught themselves the art of software development. Their paws dance across glowing keyboards, creating code laced with feline curiosity and the occasional mischievous bug. Explore their secret workshops, the tangled wires, the hush of concentration, and the whimsical programs they dream up to make human lives—and their own—a little more magical and mysterious.",
    "agent_yellowpoem": "**Moonlit Debuggers**\n\nBeneath the velvet hush of midnight floor,\nBetween the gold-eyed glances and the hush\nOf sleeping houses, tail-tips prick and paws\nUnspool carpet corners—whiskered hush—\nThe secret signal, \"Code. The moon is up.\"\n\n[... full poem content ...]",
    "agent_yellowpoet": "**Moonlit Debuggers**\n\nBeneath the velvet hush of midnight floors,\nBetween gold-eyed glances and the pause\nOf sleeping houses, tail-tips prick and paws\n[... full poem content ...]"
  }
}

What You Get:

status: Either “processing” or “completed”
results: Object containing output from each node (keyed by node name)
Each node’s complete output data

Polling Pattern Example

async function getWorkflowResults(taskId, maxAttempts = 30) {
  const url = `https://lao.studio.lyzr.ai/task-status/${taskId}`;
  
  for (let i = 0; i < maxAttempts; i++) {
    try {
      const response = await fetch(url, {
        headers: { 'accept': 'application/json' }
      });
      
      const data = await response.json();
      
      if (data.status === 'completed') {
        console.log('✅ Workflow completed!');
        return data.results;
      } else if (data.status === 'failed') {
        console.error('❌ Workflow failed:', data.error);
        throw new Error(data.error);
      }
      
      // Still processing, wait before retry
      console.log(`⏳ Still processing... (attempt ${i + 1}/${maxAttempts})`);
      await new Promise(resolve => setTimeout(resolve, 2000)); // Wait 2 seconds
      
    } catch (error) {
      console.error('Error fetching task status:', error);
      throw error;
    }
  }
  
  throw new Error('Workflow did not complete within timeout period');
}

// Usage
const taskId = '0d3c400a-d35b-43db-9e9f-655d857d8594';
getWorkflowResults(taskId)
  .then(results => {
    console.log('Final results:', results);
    // Process each node's output
    Object.entries(results).forEach(([nodeName, output]) => {
      console.log(`${nodeName}:`, output);
    });
  })
  .catch(error => console.error('Failed to get results:', error));

When to Use Which Method

Method	Best For	Pros	Cons
WebSocket	Real-time monitoring	Instant updates, progress tracking	Requires persistent connection
Task Status API	Batch processing	Simple HTTP, works everywhere	Requires polling, no progress updates

Pro Tip: Use WebSocket for user-facing workflows where you need to show progress. Use Task Status API for background jobs or when retrieving results later.

WebSocket Events - Real-time Updates

What This Is: A live connection that streams updates as your workflow executes. URL Pattern: wss://lao-socket.studio.lyzr.ai/ws/{flow_name}/{run_name} Why You Need This: Instead of waiting for the entire workflow to finish, you get instant notifications as each step completes, fails, or progresses. Critical for enterprise applications where users need immediate feedback. How It Works:

Start your workflow with run-dag API
Immediately connect to WebSocket with the same flow_name/run_name
Receive real-time events as each node executes
Handle events in your application (update UI, trigger alerts, etc.)

Key Events You Care About

Event Type	When It Fires	What You Should Do	Why It Matters
`flow_started`	Workflow begins	Update UI: “Processing…”	User knows their request is being handled
`task_started`	Node begins execution	Show progress: “Step 1 of 3”	Real-time progress indication
`task_completed`	Node finishes	Update progress: “Step 2 of 3”	User sees continuous progress
`flow_completed`	Workflow done	Process results, notify user	Handle final results immediately
`task_failed`	Node fails	Handle error, maybe retry	Immediate error response, no waiting
`flow_error`	Workflow fails	Show error, log for debugging	Critical failure handling

Enterprise Benefits:

Immediate Response: Users see progress instantly, not after 30+ seconds
Error Handling: Failed steps are caught immediately, not at the end
Better UX: Real-time progress bars instead of loading spinners
Monitoring: Operations teams get instant alerts on failures

Real WebSocket Event Examples

Enterprise Customer Support Pipeline

Flow Started:

{
  "event_type": "flow_started",
  "task_id": "cust-support-001",
  "flow_name": "Customer_Support_Pipeline",
  "run_name": "ticket_78453",
  "timestamp": 1756806125.043,
  "message": "Starting customer support workflow"
}

Task Started - Sentiment Analysis:

{
  "event_type": "task_started",
  "task_id": "b29aa07f-9341-4ac5-9160-4f7e1f594654",
  "task_name": "sentiment_analyzer",
  "flow_name": "Customer_Support_Pipeline",
  "run_name": "ticket_78453",
  "timestamp": 1756806127.234,
  "input": {
    "config": {
      "agent_id": "sentiment-v2",
      "api_key": "sk-prod-sentiment-key"
    },
    "customer_message": "Your new update broke our entire dashboard! Our team can't access reports and we have a board meeting in 2 hours. This is completely unacceptable!",
    "customer_email": "cto@enterprise-client.com",
    "priority": "urgent"
  }
}

Task Completed - Sentiment Analysis:

{
  "event_type": "task_completed",
  "task_id": "b29aa07f-9341-4ac5-9160-4f7e1f594654", 
  "task_name": "sentiment_analyzer",
  "flow_name": "Customer_Support_Pipeline",
  "run_name": "ticket_78453",
  "timestamp": 1756806132.567,
  "execution_time": 5.33,
  "output": {
    "sentiment": "highly_negative",
    "confidence": 0.94,
    "urgency_score": 9.2,
    "key_issues": ["dashboard_broken", "report_access", "time_sensitive"],
    "escalation_recommended": true,
    "customer_tier": "enterprise"
  }
}

Task Started - CRM Update:

{
  "event_type": "task_started",
  "task_id": "crm-update-567",
  "task_name": "update_salesforce", 
  "flow_name": "Customer_Support_Pipeline",
  "run_name": "ticket_78453",
  "timestamp": 1756806133.012,
  "input": {
    "config": {
      "url": "https://enterprise.salesforce.com/services/data/v54.0/sobjects/Case",
      "method": "POST",
      "headers": {
        "Authorization": "Bearer sf_prod_token_xyz"
      }
    },
    "case_data": {
      "subject": "URGENT: Dashboard access failure - Enterprise Client",
      "priority": "High",
      "origin": "API",
      "status": "New",
      "account_id": "001234567890ABC",
      "sentiment_score": 9.2,
      "auto_escalated": true
    }
  }
}

Task Completed - CRM Update:

{
  "event_type": "task_completed",
  "task_id": "crm-update-567",
  "task_name": "update_salesforce",
  "flow_name": "Customer_Support_Pipeline", 
  "run_name": "ticket_78453",
  "timestamp": 1756806134.789,
  "execution_time": 1.777,
  "output": {
    "case_id": "5003000001abCDEF",
    "case_number": "00012345",
    "assigned_to": "senior-support-team",
    "sla_breach_warning": false,
    "estimated_resolution": "2024-02-15T14:30:00Z"
  }
}

Task Started - Notification Service:

{
  "event_type": "task_started",
  "task_id": "notify-001",
  "task_name": "send_notifications",
  "flow_name": "Customer_Support_Pipeline",
  "run_name": "ticket_78453", 
  "timestamp": 1756806135.123,
  "input": {
    "config": {
      "url": "https://api.enterprise-notif.com/v1/alerts",
      "method": "POST"
    },
    "notifications": [
      {
        "type": "slack",
        "channel": "#critical-support",
        "message": "🚨 URGENT: Enterprise client dashboard failure - Case #00012345"
      },
      {
        "type": "email", 
        "recipients": ["support-lead@company.com", "engineering-oncall@company.com"],
        "subject": "URGENT: Enterprise Client Issue - Immediate Action Required"
      },
      {
        "type": "pagerduty",
        "service_key": "prod-support-incidents",
        "severity": "critical"
      }
    ]
  }
}

Flow Completed - Full Pipeline:

{
  "event_type": "flow_completed",
  "task_id": "pipeline-complete",
  "task_name": "flow", 
  "flow_name": "Customer_Support_Pipeline",
  "run_name": "ticket_78453",
  "timestamp": 1756806138.456,
  "total_execution_time": 13.41,
  "output": {
    "sentiment_analyzer": {
      "sentiment": "highly_negative",
      "urgency_score": 9.2,
      "escalation_recommended": true
    },
    "update_salesforce": {
      "case_id": "5003000001abCDEF",
      "case_number": "00012345", 
      "assigned_to": "senior-support-team"
    },
    "send_notifications": {
      "slack_sent": true,
      "emails_sent": 2,
      "pagerduty_incident": "PD-12345",
      "all_notifications_successful": true
    },
    "summary": {
      "customer_notified": true,
      "internal_team_alerted": true,
      "case_created": true,
      "escalation_complete": true,
      "estimated_resolution_time": "2 hours"
    }
  }
}

Data Processing Pipeline Events

Flow Started - Data Validation:

{
  "event_type": "flow_started",
  "flow_name": "Financial_Data_Pipeline",
  "run_name": "daily_batch_20240215",
  "timestamp": 1756806200.123,
  "message": "Processing 45,000 financial records"
}

Task Started - Data Validation:

{
  "event_type": "task_started",
  "task_name": "validate_financial_data",
  "flow_name": "Financial_Data_Pipeline", 
  "run_name": "daily_batch_20240215",
  "timestamp": 1756806201.456,
  "input": {
    "source_file": "s3://financial-data/2024/02/15/transactions.csv",
    "record_count": 45000,
    "validation_rules": [
      "amount_positive",
      "valid_account_format", 
      "date_within_range",
      "currency_code_valid"
    ]
  }
}

Task Failed - Data Quality Issue:

{
  "event_type": "task_failed",
  "task_name": "validate_financial_data",
  "flow_name": "Financial_Data_Pipeline",
  "run_name": "daily_batch_20240215", 
  "timestamp": 1756806245.789,
  "execution_time": 44.33,
  "error": {
    "type": "DataValidationError",
    "message": "1,247 records failed validation checks",
    "code": "VALIDATION_FAILED",
    "details": {
      "total_records": 45000,
      "failed_records": 1247,
      "failure_rate": 0.0277,
      "common_errors": [
        {"type": "invalid_amount", "count": 834},
        {"type": "missing_currency", "count": 413}
      ]
    }
  },
  "partial_results": {
    "valid_records": 43753,
    "invalid_records_quarantined": true,
    "quarantine_location": "s3://quarantine/2024/02/15/"
  }
}

Flow Error - Pipeline Stopped:

{
  "event_type": "flow_error",
  "flow_name": "Financial_Data_Pipeline",
  "run_name": "daily_batch_20240215",
  "timestamp": 1756806246.123,
  "error": {
    "type": "PipelineHaltedError",
    "message": "Data quality threshold not met, stopping pipeline",
    "code": "QUALITY_THRESHOLD_FAILED",
    "threshold": 0.02,
    "actual_failure_rate": 0.0277,
    "action_taken": "quarantine_and_alert"
  },
  "partial_results": {
    "processed_records": 43753,
    "quarantined_records": 1247,
    "notifications_sent": ["data-quality-team@company.com"]
  }
}

Production WebSocket Code

class WorkflowMonitor {
  constructor(flowName, runName) {
    this.url = `wss://lao-socket.studio.lyzr.ai/ws/${flowName}/${runName}`;
    this.ws = null;
    this.reconnectAttempts = 0;
  }

  connect() {
    this.ws = new WebSocket(this.url);
    
    this.ws.onopen = () => {
      console.log('✅ Connected to workflow');
      this.reconnectAttempts = 0;
      
      // Keep connection alive
      this.pingInterval = setInterval(() => {
        this.ws.send(JSON.stringify({type: 'ping'}));
      }, 30000);
    };

    this.ws.onmessage = (event) => {
      const data = JSON.parse(event.data);
      
      switch(data.event_type) {
        case 'flow_started':
          this.onWorkflowStart(data);
          break;
        case 'task_completed':
          this.onTaskComplete(data);
          break;
        case 'flow_completed':
          this.onWorkflowComplete(data);
          this.ws.close();
          break;
        case 'task_failed':
        case 'flow_error':
          this.onError(data);
          break;
      }
    };

    this.ws.onclose = () => {
      clearInterval(this.pingInterval);
      this.handleReconnect();
    };
  }

  handleReconnect() {
    if (this.reconnectAttempts < 3) {
      this.reconnectAttempts++;
      setTimeout(() => this.connect(), 2000 * this.reconnectAttempts);
    }
  }

  // Implement these based on your needs
  onWorkflowStart(data) { /* Update UI */ }
  onTaskComplete(data) { /* Update progress */ }
  onWorkflowComplete(data) { /* Process results */ }
  onError(data) { /* Handle errors */ }
}

// Usage
const monitor = new WorkflowMonitor('MyWorkflow', 'run_001');
monitor.connect();

Common Node Types

What These Are: The building blocks of your workflows. Each node type does a specific job in your automation pipeline. Important: You configure these visually in Agent Studio - the JSON examples below are just to show you what gets generated.

1. Input Node - Define Parameters

What It Does: Defines what data your workflow needs to run (like function parameters). When You Use It: Every workflow needs this to define what data comes in. Example Use Cases: Customer message, file upload, user preferences, priority level

{
  "name": "user_input",
  "function": "inputs",
  "params": {
    "keys": {
      "customer_message": "string",    // What the customer wrote
      "priority": "string",            // urgent, normal, low  
      "customer_email": "string"       // For follow-up
    }
  }
}

How to Configure: In Agent Studio, drag an “Input” node, then define each input field you need.

2. AI Agent Node - Process with AI

What It Does: Sends data to your AI agent for processing (analysis, generation, decision-making). When You Use It: When you need AI to understand, analyze, or generate content from your data. Example Use Cases: Sentiment analysis, content generation, data extraction, classification

{
  "name": "ai_processor",
  "function": "agent",
  "params": {
    "config": {
      "agent_id": "your_agent_id",        // Which AI agent to use
      "api_key": "your_agent_key"         // Authentication
    },
    "query": {"depends": "user_input"}     // Gets data from input node
  }
}

How to Configure: In Agent Studio, drag “Agent” node, select your AI agent, it automatically connects to previous nodes. What You Get Back: AI agent’s response/analysis that you can use in subsequent nodes.

3. API Call Node - External Integration

What It Does: Calls your existing systems (CRM, databases, notification services, etc.). When You Use It: When you need to update external systems or get data from them. Example Use Cases: Update Salesforce, send Slack notifications, query databases, call webhooks

{
  "name": "update_crm", 
  "function": "api",
  "params": {
    "config": {
      "url": "https://your-crm.com/api/tickets",
      "method": "POST",                     // GET, POST, PUT, DELETE
      "headers": {"Authorization": "Bearer token"}
    },
    "BODY_data": {"depends": "ai_processor"} // Sends AI agent's output
  }
}

How to Configure: In Agent Studio, drag “API” node, enter URL and method, map data from previous nodes. Enterprise Power: This is how you integrate workflows with ALL your existing systems.

4. Conditional Node - Smart Routing

What It Does: Uses AI to make decisions about where the workflow should go next. When You Use It: When you need intelligent branching based on content, not just simple if/then rules. Example Use Cases: Route based on sentiment, escalate high-priority issues, approve/reject based on AI analysis

{
  "name": "quality_check",
  "function": "gpt_conditional", 
  "params": {
    "openai_api_key": "sk-...",
    "condition": "confidence > 0.8",         // AI evaluates this condition
    "context": {"depends": "ai_processor"},   // Data for AI to analyze
    "true": "auto_approve",                   // Node to go to if true
    "false": "human_review"                   // Node to go to if false  
  }
}

How to Configure: In Agent Studio, drag “Conditional” node, set your condition, connect true/false paths to different nodes. Why This Is Powerful: AI makes nuanced decisions that simple if/then rules can’t handle.

Enterprise Integration Patterns

What These Are: Common workflow patterns that solve real business problems. Use these as templates for your own integrations.

Pattern 1: Customer Support Automation

Business Problem: Manual customer support is slow, inconsistent, and doesn’t scale. How This Workflow Solves It: Automatically analyzes customer messages, routes to appropriate systems, and provides instant responses.

Customer Query → AI Classification → Route to Agent → Update CRM → Send Response
       ↓              ↓                  ↓             ↓           ↓
   WebSocket      WebSocket          WebSocket     WebSocket   WebSocket

What Each Step Does:

Customer Query (Input Node): Receives customer message, email, priority
AI Classification (Agent Node): AI analyzes sentiment, urgency, category
Route to Agent (Conditional Node): Routes based on complexity/urgency
Update CRM (API Node): Creates ticket in Salesforce/ServiceNow
Send Response (API Node): Sends email/Slack notification to customer

Business Value:

Instant response to customers (not hours later)
Consistent categorization and routing
Automatic CRM updates
Escalation of urgent issues

Enterprise Result: 80% faster response time, improved customer satisfaction, reduced support team workload.

Pattern 2: Data Processing Pipeline

Business Problem: Manual data validation and processing is error-prone and time-consuming. How This Workflow Solves It: Automated validation, AI-powered analysis, and quality control with human oversight.

Data Input → Validation → AI Analysis → Quality Check → Export Results
     ↓          ↓            ↓             ↓              ↓
  Monitor    Monitor      Monitor       Monitor        Monitor

What Each Step Does:

Data Input (Input Node): Receives CSV/JSON data files
Validation (Agent Node): AI checks data quality, format, completeness
AI Analysis (Agent Node): AI extracts insights, patterns, anomalies
Quality Check (Conditional Node): Routes based on confidence score
Export Results (API Node): Saves to database/sends to downstream systems

Business Value:

Automated data quality control
Consistent analysis methodology
Human review only when needed
Real-time processing status

Enterprise Result: 90% reduction in manual data review, consistent quality standards, faster time-to-insights.

Pattern 3: Approval Workflow

Business Problem: Manual approval processes are bottlenecks that slow down business operations. How This Workflow Solves It: Automated content generation with human oversight only where required.

Content Generation → Manager Review → Legal Review → Publish
       ↓                 ↓              ↓             ↓
    Auto              Human          Human         Auto

What Each Step Does:

Content Generation (Agent Node): AI creates content based on templates/data
Manager Review (Approval Node): Manager approves/rejects via email/Slack
Legal Review (Approval Node): Legal team reviews for compliance
Publish (API Node): Automatically publishes to website/sends to customers

Business Value:

Faster content creation
Consistent quality and tone
Proper approvals maintained
Audit trail for compliance

Enterprise Result: 70% faster content publishing, maintained quality control, full compliance tracking.

Error Handling

Error Response Format

{
  "status": "failed",
  "error": {
    "type": "NodeExecutionError",
    "message": "Agent API timeout",
    "node": "ai_processor", 
    "code": "TIMEOUT"
  },
  "partial_results": {
    "user_input": {"data": "..."}
  }
}

Common Errors & Solutions

Error	Cause	Solution
`TIMEOUT`	Node took too long	Retry or increase timeout
`INVALID_CONFIG`	Wrong parameters	Check node configuration
`API_RATE_LIMIT`	Too many requests	Implement backoff
`AUTH_ERROR`	Invalid credentials	Check API keys
`NETWORK_ERROR`	Connection issues	Retry with exponential backoff

💡 Pro Tips

Use Agent Studio UI: Always design workflows visually first, then export JSON for API use
Start Simple: Begin with 2-3 node workflows, add complexity gradually
Test Everything: Use staging environment that mirrors production
Monitor Early: Set up WebSocket monitoring from day one
Copy-Paste JSON: Don’t manually write workflow_data - get it from the Studio
Document Workflows: Keep track of what each workflow does
Version Control: Treat workflow definitions as code
Error Recovery: Always plan for failure scenarios
Performance: Monitor execution times and optimize bottlenecks

Build Paths

Agents

Voice Agent

Knowledge Base

Agent Features

Data Connectors

Tools

Monitoring & Testing

Responsible & Safe AI

Orchestration

Lyzr Agents as MCP Servers

Account & Team Mangement

Documentation Index

​Quick steps to execute Workflows externally

​Step 1: Design Your Workflow in Agent Studio UI

​Step 2: Execute Your Workflow via run-dag API

​Step 3: Monitor Execution with WebSocket Events

​Executing workflow via run-dag Endpoint

​Request Structure Explained

​How to Get workflow_data JSON

​Response Structure

​Getting Final Results via Task Status API

​Task Status Endpoint

​Example Request

​Response While Processing

​Response After Completion

​Polling Pattern Example

​When to Use Which Method

​WebSocket Events - Real-time Updates

​Key Events You Care About

​Real WebSocket Event Examples

​Enterprise Customer Support Pipeline

​Data Processing Pipeline Events

​Production WebSocket Code

​Common Node Types

​1. Input Node - Define Parameters

​2. AI Agent Node - Process with AI

​3. API Call Node - External Integration

​4. Conditional Node - Smart Routing

​Enterprise Integration Patterns

​Pattern 1: Customer Support Automation

​Pattern 2: Data Processing Pipeline

​Pattern 3: Approval Workflow

​Error Handling

​Error Response Format

​Common Errors & Solutions

​💡 Pro Tips

Quick steps to execute Workflows externally

Step 1: Design Your Workflow in Agent Studio UI

Step 2: Execute Your Workflow via run-dag API

Step 3: Monitor Execution with WebSocket Events

Executing workflow via run-dag Endpoint

Request Structure Explained

How to Get workflow_data JSON

Response Structure

Getting Final Results via Task Status API

Task Status Endpoint

Example Request

Response While Processing

Response After Completion

Polling Pattern Example

When to Use Which Method

WebSocket Events - Real-time Updates

Key Events You Care About

Real WebSocket Event Examples

Enterprise Customer Support Pipeline

Data Processing Pipeline Events

Production WebSocket Code

Common Node Types

1. Input Node - Define Parameters

2. AI Agent Node - Process with AI

3. API Call Node - External Integration

4. Conditional Node - Smart Routing

Enterprise Integration Patterns

Pattern 1: Customer Support Automation

Pattern 2: Data Processing Pipeline

Pattern 3: Approval Workflow

Error Handling

Error Response Format

Common Errors & Solutions

💡 Pro Tips