ActivePieces Integration

Mazaal AI's automation platform is built on a powerful fork of ActivePieces, enhanced with seamless AI agent capabilities. This integration gives you the best of both worlds: a robust, open-source workflow automation foundation with enterprise-grade AI features.

What is ActivePieces?

ActivePieces is an open-source workflow automation platform that allows you to automate tasks across different applications without coding. Mazaal AI has taken this powerful foundation and enhanced it with our AI agent technology to create a uniquely powerful business automation solution.

🔄 Enhanced ActivePieces: While our platform is based on ActivePieces, we've significantly extended its capabilities with AI integration, additional connectors, and enterprise features. We refer to our implementation as the "Automation Platform" to reflect these enhancements.

How Mazaal AI Enhances ActivePieces

AI Agent Integration

The most significant enhancement we've made to ActivePieces is seamless integration with Mazaal AI agents:

• AI Agent Pieces: Use your trained AI agents directly within workflows
• Content Analysis: Analyze text, documents, and images with AI
• Dynamic Response Generation: Create personalized content based on context
• Decision Making: Use AI to make intelligent routing decisions
• Data Extraction: Pull structured data from unstructured content

These capabilities transform ActivePieces from a simple automation tool into an intelligent business process platform.

Extended Connector Library

We've expanded the ActivePieces connector library with additional integrations:

• Industry-Specific Tools: Connectors for healthcare, finance, legal, and other specialized industries
• Enterprise Systems: Expanded integration with ERP, CRM, and HRIS platforms
• Data Processing: Advanced data transformation and analysis tools
• Custom Connectors: Ability to create and share custom connectors

Our platform now includes 230+ pre-built connectors, compared to the 100+ in the open-source version.

Enterprise Capabilities

We've added enterprise-grade features essential for business use:

• Advanced Security: Role-based access control, audit logs, and encryption
• Team Collaboration: Shared workflows, comments, and version control
• Monitoring & Alerts: Performance tracking and failure notifications
• High Availability: Redundant infrastructure for mission-critical workflows
• Compliance Features: Tools to help maintain regulatory compliance

These enhancements make our platform suitable for organizations of all sizes, from startups to enterprises.

Improved User Experience

We've refined the user interface and experience:

• Simplified Builder: More intuitive workflow creation
• Templates Gallery: Pre-built workflows for common scenarios
• Guided Setup: Step-by-step assistance for new users
• Testing Tools: Enhanced debugging and testing capabilities
• Mobile Optimization: Better experience on tablets and smartphones

These improvements make the platform more accessible to users of all technical levels.

Key Components of the Platform

• 🔧 Flow Builder: Visual interface for creating and editing workflows with drag-and-drop simplicity
• 🎯 Triggers: Events that start workflows, including webhooks, schedules, and app-specific events
• ⚙️ Actions: Tasks that workflows perform, from sending emails to updating records
• 🔌 Connections: Integrations with external services and applications
• 🧠 Logic: Conditional branching, loops, and other flow control elements
• ✨ AI Pieces: Special components that leverage Mazaal AI capabilities

Using AI Pieces in Your Workflows

The most powerful aspect of our ActivePieces integration is the ability to use AI agents directly within your workflows. Here's how to leverage this capability:

1. Add an AI Piece to Your Workflow In the Flow Builder, click the "+" button to add a new step, then select "Mazaal AI" from the apps list.

   

2. Choose the AI Action Select from available AI actions:

   • Query Agent: Ask your AI agent a question
   • Analyze Text: Perform sentiment analysis, categorization, or entity extraction
   • Generate Content: Create text based on parameters
   • Extract Data: Pull structured data from unstructured text
   • Translate: Convert text between languages

3. Configure the AI Piece For a "Query Agent" action:

   1. Select which of your trained agents to use
   2. Define the query (can use dynamic data from previous steps)
   3. Set additional context parameters if needed
   4. Configure how to handle the response

4. Use the AI Output in Subsequent Steps The AI piece will return structured data that can be used in later steps of your workflow:

   • The main response text
   • Confidence score
   • Detected entities or categories
   • Extracted data points
   • Recommended actions

5. Test Your AI Integration Use the testing tools to run your workflow and verify that the AI piece is working as expected.

Real-World Example: AI-Powered Document Processing

💬 "We used to have three team members spending hours every day manually processing invoices. Now our Mazaal AI workflow extracts the data automatically, validates it against our systems, and only involves humans for exceptions. What took hours now happens in seconds." — Financial Operations Director at a manufacturing company

The Challenge

A manufacturing company was struggling with their accounts payable process:

• Receiving 100+ supplier invoices daily in various formats (PDF, email, scanned documents)
• Manually extracting key information (invoice number, date, amount, line items)
• Entering data into their accounting system
• Matching invoices to purchase orders
• Routing for approval based on amount and department

The process was time-consuming, error-prone, and created a bottleneck in their financial operations.

The Solution

They created a Mazaal AI workflow that:

1. Monitored a dedicated email inbox for incoming invoices
2. Used AI to extract text from PDF and image attachments
3. Applied an AI agent trained on invoice processing to extract structured data:
   • Vendor information
   • Invoice number and date
   • Line items and amounts
   • Payment terms
4. Validated the extracted data against their systems
5. Matched invoices to purchase orders automatically
6. Created records in their accounting system
7. Routed for approval based on business rules
8. Flagged exceptions for human review

The Results

After implementing their AI-powered workflow:

• Processing time reduced from 15 minutes to 30 seconds per invoice
• Data entry errors reduced by 94%
• Staff reallocated to higher-value financial analysis
• Faster payment processing led to better supplier relationships
• Early payment discounts captured, saving thousands monthly
• Complete audit trail for compliance purposes

The ROI was achieved in less than two months, and the system continues to improve as the AI learns from exceptions.

Advanced ActivePieces Features

As you become more comfortable with the platform, explore these advanced capabilities:

Custom JavaScript Functions

For specialized logic that can't be handled by standard pieces:

// Example: Custom data transformation
function transformData(inputData) {
  const result = {};
  
  // Extract first and last name from full name
  const nameParts = inputData.fullName.split(' ');
  result.firstName = nameParts[0];
  result.lastName = nameParts.slice(1).join(' ');
  
  // Format phone number
  result.formattedPhone = inputData.phone.replace(/(\d{3})(\d{3})(\d{4})/, '($1) $2-$3');
  
  // Calculate age from birthdate
  const birthDate = new Date(inputData.birthdate);
  const today = new Date();
  result.age = today.getFullYear() - birthDate.getFullYear();
  
  return result;
}

Nested Workflows

Create modular, reusable workflow components:

• Workflow Structure/
  • Main Order Processing Workflow
  • Sub-Workflows/
    • Customer Validation
    • Inventory Check
    • Payment Processing
    • Shipping Arrangement

This approach improves maintainability and allows specialized team members to focus on their areas of expertise.

Webhook API Creation

Create custom APIs by exposing workflows as webhooks:

Endpoint: https://api.mazaal.ai/webhook/your-unique-id
Method: POST
Authentication: API Key
Request Body: JSON with required parameters
Response: Result of the workflow execution

This allows your workflows to be triggered by external systems or custom applications.

Error Handling Strategies

Implement robust error handling for production workflows:

Retry Logic Configuration

For transient errors like network issues:

• Set maximum retry attempts (e.g., 3-5 times)
• Configure increasing delay between retries (exponential backoff)
• Define conditions for retry vs. failure

Example: A payment gateway connection might fail temporarily but succeed on retry.

Alternative Path Definition

For workflows with critical functions:

• Create backup paths for when primary methods fail
• Define fallback services or APIs
• Implement graceful degradation of functionality

Example: If your primary email service is down, route notifications through a secondary provider or SMS.

Notification System

For critical workflow failures:

• Send alerts to responsible team members
• Include detailed error information
• Provide links to monitoring dashboards
• Escalate based on severity and time without resolution

Example: Critical payment processing errors might trigger immediate SMS alerts to the finance team.

Comprehensive Error Logging

For troubleshooting and improvement:

• Log detailed error information
• Capture input data (sanitized of sensitive information)
• Record system state at time of failure
• Track error patterns over time

Example: Analyzing error logs might reveal that a particular vendor's invoices consistently cause extraction failures, indicating a need for specialized training.

Differences from Open-Source ActivePieces

While our platform is built on ActivePieces, there are some key differences to be aware of:

Warning: If you're familiar with the open-source version of ActivePieces, note these differences when using Mazaal AI's implementation.

Feature Comparison

API and Configuration Differences

If you're migrating from self-hosted ActivePieces:

• Our authentication system uses a different structure
• Some connector configurations have additional options
• Our rate limits and quotas are based on your subscription plan
• We've enhanced the webhook functionality with additional security features

Getting Started with the Platform

If you're new to workflow automation, follow these steps to get started:

1. Explore the Templates Gallery Browse our pre-built workflow templates to find examples similar to your needs.

2. Create Your First Simple Workflow Start with a basic workflow, such as:

   • Form submission to email notification
   • Scheduled social media posting
   • New lead to CRM creation

3. Add AI Capabilities Once comfortable with basic workflows, add AI pieces to enhance functionality:

   • Use AI to categorize incoming requests
   • Generate personalized responses
   • Extract data from unstructured content

4. Connect Your Business Systems Integrate with your existing tools:

   • CRM systems
   • Marketing platforms
   • Accounting software
   • Communication tools

5. Monitor and Optimize After deployment, regularly review performance:

   • Check execution logs
   • Identify bottlenecks or failures
   • Refine based on real-world usage

Resources and Support

To help you make the most of our ActivePieces-based platform:

• 📚 Documentation: Detailed guides and API references
• 🎥 Video Tutorials: Step-by-step visual instructions
• 👥 Community Forum: Connect with other users
• 🧑‍💻 Support Team: Get help from our experts

✨ AI + Automation Power: The combination of ActivePieces' powerful workflow engine and Mazaal AI's intelligent agents creates a platform that can transform your business processes—automating routine tasks while handling complex, unstructured information with human-like understanding.