Welcome to the ZBrain documentation!

This comprehensive resource is designed to help you explore ZBrain, a unified platform for enterprise AI enablement that guides organizations through each stage of their AI journey, from transforming internal workflow challenges into structured AI use cases to evaluating readiness, identifying opportunities, and implementing full-scale solutions. The documentation covers the platform’s key components and demonstrates how ZBrain can deliver value across your organization.

ZBrain consists of three core platforms: ZBrain XPLR, for discovering and planning AI solutions across the enterprise, ZBrain Builder, for developing, deploying, and managing those solutions efficiently and ZBrain CoI, for enabling structured ideation and collaboration to transform operational challenges into investment-ready AI use cases. It ensures your enterprise has a strategic roadmap for AI adoption tailored to its unique workflows and objectives.

Within this guide, you will discover how to leverage ZBrain Builder to create and manage knowledge bases, import data from multiple sources, design automated workflows, develop AI-powered applications and agents, configure prompts for customized AI behavior, and deploy scalable, enterprise-grade AI solutions. Also, you'll learn how to use ZBrain CoI to accelerate AI opportunity discovery by submitting your process-level challenges via its prompt-driven interface. CoI converts these submissions into detailed AI use case reports that include cost estimates, business impact, and ROI metrics. All activities take place within a secure, multi-tenant environment that supports editing, sharing, status tracking, and alignment with organizational objectives.

Whether you are initiating your AI journey or looking to scale existing initiatives, this documentation serves as your go-to resource for building and managing intelligent, enterprise-ready AI solutions with ZBrain.

Knowledge base

ZBrain’s knowledge base is a centralized data repository that enables AI applications and agents to access domain-specific information for accurate and context-aware responses. It supports a wide range of data formats, including documents, images, and videos. You can also import data from sources like Web URLs, Webhooks, Google Sheets, Google Slides, Google Docs, ElasticSearch, Notion, MongoDB, ServiceNow, Confluence, Jira and more. This broad data connectivity ensures flexibility and scalability, allowing your AI systems to stay informed and deliver precise, domain-relevant outputs.

ZBrain Builder is an all-in-one orchestration platform for enterprise-grade AI solutions that enables organizations to quickly develop, deploy, and manage intelligent custom GenAI agents and applications without requiring extensive developer resources. The platform is designed to democratize AI technology, making it accessible and scalable for organizations of all sizes. ZBrain Builder integrates seamlessly with your existing technology stack, acting as a central hub for all your GenAI-based solutions. As an enterprise-grade agentic AI platform, it offers all the necessary tools to build custom generative AI solutions that efficiently handle a wide array of tasks. It goes beyond traditional NLP tasks, offering end-to-end orchestration of AI agents capable of handling complex workflows across functions like legal, HR, finance, sales, marketing, and IT. With built-in data governance, secure APIs, smart orchestration logic, and continuous learning via human feedback, ZBrain empowers enterprises to drive operational excellence at scale.

Key features of ZBrain Builder

• Advanced knowledge base: ZBrain Builder ingests data from various sources (documents, web URLs, databases) in multiple formats (PDF, TXT, CSV, JSON) and optimizes it at the chunk level for efficient retrieval. It supports various vector stores and is agnostic to the underlying storage provider. This rich knowledge base fuels all ZBrain agents and applications, ensuring they have the information they need to deliver accurate results.

• Low-code development with Flow: ZBrain Builder's Flow feature offers pre-built components for the rapid development of complex AI applications. The intuitive interface enables the integration of content from various sources, real-time data fetching, and the incorporation of third-party tools, applications, and conditional logic into workflows, with webhook triggers to build your app’s intricate business logic.

• Human in the loop: The platform gathers feedback from end-users on AI outputs and performance, allowing operators to provide corrections and guidance to improve AI model efficiency. This process helps refine the models' output and optimize data retrieval based on human input.

• Extended database: ZBrain Builder allows operators to extend their data at the chunk or file level with additional information and update the meta-information associated with data entries. It also offers data summarization and ontology generation capabilities.

• Cloud and model agnostic: ZBrain solutions can be deployed on-premise or in the cloud and interact seamlessly with major LLMs. It supports OpenAI’s GPT-4, Anthropic’s Claude, Google’s Gemini, Meta’s Llama-3, and other proprietary or open-source models​ like Gemma and Mistral.

• Evaluation suite of tools: ZBrain Builder provides automated test suites for continuous validation of AI agents and applications, implements guardrails to ensure output quality, and uses LLMs to assess app results.

• Advanced prompting techniques: It supports advanced prompting techniques, including Zero/Few Shot Prompting, Chain of Thought Prompting, Self Consistency, Retrieval Augmentation Generation, Self Reflection, and Automatic Prompt Engineering. These techniques ensure the highest accuracy in results, delivering robust and reliable outputs for your applications.

• APPOps: ZBrain Builder keeps your AI applications running smoothly with its built-in application operations (APPOps) features. It monitors application health and performance by proactively conducting continuous background validation and identifying and resolving issues before they impact users. This ensures your AI solutions are reliable and deliver consistent results.

• Agents: ZBrain Builder enables you to create AI agents that automate complex business processes across enterprise functions. The agents can be configured according to specific needs and seamlessly integrate into existing workflows, optimizing repetitive and decision-intensive tasks. With advanced capabilities, AI agents built on ZBrain Builder handle complex processes, provide precise insights, reduce manual effort, and boost productivity, enabling teams to focus on strategic goals and drive operational efficiency and agility. An agent store is also available that allows users to quickly select and deploy pre-built specialized agents, with support for multi-agent coordination to handle complex workflows.

How ZBrain Builder works

ZBrain Builder integrates data from both private and public sources into its advanced knowledge base, enabling efficient retrievals and enhanced data management through extended databases. Its powerful engine handles critical business logic, data, and user governance, and runtime integrations. ZBrain Builder facilitates low-code agent development through its agents layer, which includes agent collaboration, prebuilt agent library, feedback loop and retrieval optimization. The interface layer provides seamless connectivity via APIs, SDKs, and user-friendly interfaces, enabling easy integration with existing systems. ZBrain Builder supports private enterprise deployment, offering integration with both proprietary and open-source large language models (LLMs).

Data sources

ZBrain Builder ingests and processes multimodal data, including text, audio, images, and video, from both private and public sources:

A. Private data

These sources contain organization-specific and confidential company information, which is securely accessed for processing. The following sources provide access to private organizational data:

• Enterprise apps: Connectors to systems like Salesforce, SAP, and other CRM/ERP platforms.

• Business systems: Integrations with core business tools such as Microsoft Office, PDF documents, and Google Workspace, enabling structured and semi-structured data handling.

• Data clouds: Seamless connections with data lake and warehouse solutions like Snowflake and Databricks, enabling high-volume, scalable analytics.

B. Public data

ZBrain Builder taps into publicly accessible information from sources such as Google, Yahoo, Bing, and Wikipedia, enriching enterprise insights with external knowledge.

ZBrain Builder: The orchestration layer

At the core of the platform, the ZBrain Builder enables configuration, orchestration, and management of intelligent workflows, AI agents, and system behavior across the following key layers:

A. Knowledge base

ZBrain Builder has a dynamic and structured knowledge layer for intelligent querying:

• Extended DB: Allows expansion and enrichment of the database with additional information for more accurate results.

• Retrievals: Indexes and surfaces relevant documents, records, and entities efficiently for fast data retrieval.

• Automated reasoning: Extracts and applies rules and variables from the knowledge base to deliver precise, logic-driven answers to user queries.

B. Builder engine

This is the core processing engine enabling:

• Business logic & OOTB algorithms: Manages core business rules while providing out-of-the-box algorithms for various functions.

• Data & user governance: Enforces secure data handling, access controls, and auditability across users and roles for robust data security and governance.

• Runtime integrations: Ensures agents stay connected to live systems and databases for real-time task execution.

C. Agents

The agent layer is the operational backbone where AI actions are defined and executed:

• Low-code platform: Simplifies the development of agents with minimal coding, accelerating time-to-market.

• Memory: Agents can retain contextual understanding for ongoing tasks or long-term sessions.

• Autonomous agents: Execute complex, multi-step tasks through autonomous reasoning and decision-making, enabling end-to-end workflow automation without manual intervention.

• Agent collaboration

D. Tuning layer

The tuning layer focuses on AI solutions' performance and safety. It consists of:

• Monitoring: Tracks system performance and model usage in real-time.

• Evaluation suite: Provides necessary tools to benchmark agent accuracy continuously and LLM output quality.

• Self-improvement: Facilitates autonomous refinement of agents and model prompts over time.

• Guardrails: Applies predefined rules and guidelines to guide AI behavior and prevent harmful or biased responses.

Interface layer

ZBrain Builder offers flexible interfaces for both developers and business users. Its interface layer consists of:

• APIs: Programmatic interfaces that enable the smooth integration of ZBrain solutions with other enterprise systems.

• SDKs: Prebuilt libraries and kits help accelerate the custom development and integration of agents and solutions.

• User interface: Intuitive front-end interface allowing users to interact with ZBrain's AI-powered solutions easily.

LLM layer

ZBrain Builder is model-agnostic and supports flexible large language model configurations through:

• Private models: Supports seamless integration with private instances of major AI platforms like Google Vertex AI, Amazon Bedrock, Azure OpenAI Service, and Together AI.

• Model as a service: Enables integration with cloud-based AI services such as Hugging Face and Groq, enabling access to cutting-edge models.

• Proprietary models: Includes specialized, custom models like Microsoft’s Phi-3 and Google’s PaLM 2 for specific use cases.

Cloud compute layer

The cloud compute layer enables ZBrain Builder to harness scalable cloud-based services for tasks like search and document processing, boosting the platform's overall capabilities.

• Vector databases: Integration with advanced vector search technologies such as Pinecone, AWS OpenSearch, Vertex AI Vector Search, and Azure AI Search.

• Document processing: Support for AI-driven document processing using tools like Amazon Textract, Azure AI Vision, and Google Document AI.

Deployment & integration

• Private cloud deployment: ZBrain apps and agents can be deployed in private cloud environments for added security and control.

• Enterprise private deployment: ZBrain apps and agents can be deployed on major cloud providers like AWS, Google Cloud, and Azure, ensuring full data privacy and security while meeting enterprise-specific needs.

Why use ZBrain Builder?

• Enterprise-ready: Designed for secure, private deployment in enterprise environments.

• Efficiency: Streamlines processes through automated data collection, analysis, and content generation.

• Customization: Tailors AI solutions to specific business needs, integrating seamlessly with existing workflows.

• Operational enhancement: Improves decision-making, operational efficiency, and customer experience.

Get started with ZBrain Builder by signing up for a new account or signing in to your existing account to create, customize, and deploy AI apps and agents tailored to your organization’s needs.

New user sign-up process

ZBrain Builder offers flexible registration options to accommodate different user preferences and organizational requirements. You can create an account using either your email credentials or Google authentication.

The Apollo piece in ZBrain Flow allows you to integrate person and company enrichment features from Apollo into your workflows. With this connector, you can access Apollo’s B2B data to automatically match individual profiles or enhance company information using minimal inputs like name, email, domain, or LinkedIn URL. This is useful for lead scoring, CRM enrichment, sales automation, or any workflow where having accurate and enriched business data helps improve decisions or drive engagement.

How to Use Apollo in ZBrain Flow

Select Apollo as Your Action

1. In ZBrain Flow, click the + button and search for Apollo.

2. Select the action you need depending on your use case. ZBrain Flow supports the following Apollo actions:

   • Match Person – Retrieve detailed person-level data based on attributes like email.

   • Enrich Company – Get enriched information about a company using a domain name.

Set Up the Apollo Connection

Before using any Apollo actions, you must connect your Apollo account. To create the connection:

1. In the ‘Connection’ dropdown, click ‘Create Connection’.

2. In the pop-up window, fill in the following fields:

   • Connection Name: A name to identify this connection (e.g., Apollo)

   • API Key: Your Apollo API key

How to Use Apollo to Match a Person in ZBrain Flow

• Follow the connection setup steps provided in the “Set Up the Apollo Connection” section above.

• In the 'Email' field, enter the email address of the person you want to match.

• Optionally, enable 'Cache Response' to store the result in the project store for later use. This helps avoid repeated API calls for the same input.

Once configured, the action will return matched person data from Apollo based on the provided email.

How to Use Apollo to Enrich Company Data in ZBrain Flow

• Follow the connection setup steps provided in the “Set Up the Apollo Connection” section above.

• Enter the company’s domain in the 'Domain' field.

• You can enable 'Cache Response' to store the result in the project store and reuse the data across other steps without making a repeat API call.

After setup, this action returns enriched company data pulled from Apollo using the specified domain.

Managing knowledge base settings and configurations

The Knowledge Base Settings allow you to view and manage key configurations of your knowledge base. The name and description can be edited at any time, while other settings reflect your initial setup choices and are shown here for reference.

• Knowledge Base Name

Amazon Textract is an AWS service that automatically extracts text, handwriting, and data from scanned documents. It goes beyond simple optical character recognition (OCR) to identify, understand, and extract data from forms and tables.

How to Connect Amazon Textract to ZBrain Flow?

Step 1: Select Amazon Textract as Your Connection

How to create a summary of the file?

1. Click on the ‘File Summary’ tab.

The Azure Communication Services piece in ZBrain Flow provides integration with Microsoft's cloud communication platform, allowing you to send emails directly from your workflows. This connector enables you to automate email communications without manual intervention. This piece is ideal for organizations looking to streamline their email processes, automate routine communications, and ensure reliable message delivery as part of their broader business workflows.

How to Use Azure Communication Services in ZBrain Flow

Configuration Steps:

Data Mapper is a tool in ZBrain Flow designed to manipulate data structures by transforming and mapping data from one format to another. It helps streamline data processing, ensuring seamless integration between different systems and workflows.

To integrate Data Mapper with ZBrain Flow, follow these detailed steps:

• Click the '+' button in the Flow and search for Data Mapper in ZBrain Flow.

• Select the specific task from the list you want to include in the Flow.

Hacker News Component

The Hacker News is a popular cybersecurity publication for IT professionals seeking actionable insights. This component retrieves the latest top stories from Hacker News.

The Binance piece in ZBrain Flow provides integration with Binance's cryptocurrency exchange platform, allowing you to access real-time market data directly from your workflows. This connector enables you to retrieve current cryptocurrency prices and trading information without requiring manual checks or custom API code. With Binance integration, you can build automations that monitor crypto markets, perform calculations using current rates, and incorporate live market data into your business processes. This piece is ideal for financial workflows, portfolio management, and any automation that requires up-to-date cryptocurrency information.

How to Use Binance in ZBrain Flow?

Step 1: Select Binance as Your Connection

The Bonjoro piece in ZBrain Flow provides integration with Bonjoro's personalized video messaging platform, allowing you to automate the creation of personalized video tasks within your workflows. This connector enables you to set up video greeting tasks for your team members to record and send to customers, prospects, or team members. This piece is ideal for businesses looking to add a personal touch to their communication processes and strengthen relationships through authentic video interactions.

How to Use Bonjoro in ZBrain Flow?

Step 1: Select Bonjoro as Your Connection

• Click on the '+' button in the Flow and search for Bonjoro.

• Select Bonjoro.

• Decide on the action you need, then select it. ZBrain Flow provides two options:

  • Create a Greet – Generate a new greet in Bonjoro.

  • Custom API Call – Send a tailored API request to a specified endpoint.

How to Generate a New Greet in Bonjoro

Step 1: Connect to Bonjoro

• From the ‘Connection' dropdown, click on 'Create connection.’

• When creating a new connection, a new window appears where you give a unique name to your connection and provide your Bonjoro API credentials. Then click ‘Save.’

Step 2: Define Task Details

• In the 'Note' field, enter instructions or context for the person recording the video.

• In the 'Email' field, enter the recipient's email address where the video will be sent.

• In the 'First Name' field, enter the recipient's first name for personalization.

• In the 'Last Name' field, optionally enter the recipient's last name.

Step 3: Assign and Categorize

• Select the assignee in the ‘Assignee’ field.

• Choose the campaign from the ‘Campaign’ dropdown.

• From the 'Template' dropdown, select which message template to use for the video email.

Step 4: Add Custom Data (Optional)

• In the 'Custom Attributes' field, add any additional JSON data to include with the task.

How to Send a Tailored API Request to a Specified Endpoint

Step 1: Connect to Bonjoro

• To connect to your Bonjoro account, follow the first step in the "How to Generate a New Greet in Bonjoro" section above.

Step 2: Specify the API Endpoint

• In the 'URL' field, enter the Bonjoro API base URL with the specific endpoint you want to access.

Step 3: Select Request Method

• From the 'Method' dropdown, choose the appropriate HTTP method (GET, POST, PUT, DELETE, etc.). Different endpoints require different methods according to the Bonjoro API documentation.

Step 4: Configure Headers (Optional)

• Click 'Add Item' in the 'Headers' section to add any custom headers. Note that authorization headers are automatically injected from your connection.

Step 5: Add Query Parameters (Optional)

• Click 'Add Item' in the 'Query Parameters' section to add URL parameters. These are useful for filtering, pagination, or specific data requests.

Step 6: Provide Request Body (Optional)

• For POST, PUT, or PATCH requests, enter the JSON data to send. This should follow the structure specified in the Bonjoro API documentation.

Step 7: Set Error Handling (Optional)

• Enable 'No Error on Failure' if you want the flow to continue even if the API request fails.

• Set a 'Timeout' value in seconds to limit the request duration.

CSV (Comma-separated Values) is a simple file format used to store tabular data, such as a spreadsheet or database. In a CSV file, each line of text represents a row of data, and values within each row are separated by commas (or sometimes other delimiters like semicolons or tabs).

How to Connect CSV to ZBrain Flow?

Step 1: Select CSV as Your Connection

• Click on the '+' button in the Flow and search for CSV.

• Select CSV.

• Decide on the action you need. ZBrain Flow provides two options:

  • Convert CSV to JSON – Converts a CSV string into a JSON array.

  • Convert JSON to CSV – Converts a JSON array into CSV format.

How to convert CSV to JSON?

Step 1: Enter Your CSV Data

In the ‘CSV Text’ field, paste or input your CSV content. This is the CSV data you want to convert into JSON format.

Step 2: Configure Headers Setting

Select whether your CSV data includes headers in the ‘Does the CSV have headers?’ dropdown.

Step 3: Specify the Delimiter Type

From the ‘Delimiter Type’ dropdown, select the character that separates values in your CSV:

• Comma (,) - most common CSV format

• Tab - tab-separated values

How to convert JSON to CSV?

Step 1: Enter Your JSON Array

In the ‘JSON Array’ field, input your JSON data. The input must be a valid JSON array containing objects.

Step 2: Specify the Delimiter Type

From the ‘Delimiter Type’ dropdown, select the character that will separate values in your CSV output:

• Comma - standard CSV format using commas to separate values

• Tab - tab-separated values

Google Search is a web search engine provided by Google. It enables users to quickly retrieve relevant information from billions of web pages by indexing and ranking content based on user queries, making it ideal for research, information discovery, and accessing the latest online content.

How to Configure Google Search to ZBrain Flow?

To integrate Google Search with ZBrain Flow, follow these detailed steps:

• Click the '+' button in the Flow and search for Google Search in the ZBrain Flow.

• Provide the API Key value. Get your API Key from Settings > My Account in your ZBrain account. The authentication key is required to connect to the Google Search or SERP API. It verifies your account and grants access to the search service.

• To create a connection, click on the Create a new connection from the Connection field.

• Provide a connection name in the Connection Name field.

• Provide the API Key value.

• Save the connection.

• Enter the query to search for in the Search Query field. The term or phrase you want to look up. This field defines what information you’re trying to retrieve from Google Search.

• Select the country from the Country drop-down to specify the country or region to tailor search results. This helps you get more localized or relevant content.

• Select the search type you want from the Search Type drop-down. It defines the category of results to retrieve (e.g., web pages, images, or news). Selecting a specific type narrows the search to that content format.

• Additional query parameters you can add to refine or filter search results (e.g., language settings, safe search options) by providing values in Google Search Parameters. This ensures the returned data aligns more closely with your requirements.

Retrieval testing is used to identify similar chunks in response to a query, determining which data chunks will be processed by the language model. Users can submit queries related to the data stored in the knowledge base. The system then retrieves relevant chunks of information, ranking them based on how well they match the user's query. The most relevant chunk, with the highest matching score, is displayed first, followed by chunks with lower scores.

By analyzing the returned chunks, users can assess how well the LLM interprets and retrieves relevant information for various queries.

1. To perform retrieval testing, navigate to the ‘Retrieval Testing’ tab. Enter a query related to your data in the provided text box, and click the ‘Test’ button to execute the test.

1. You will be provided with a retrieval result, which is an array of chunks from the knowledge base.

1. You can find a minimum matching score for chunks to filter search results based on relevance.

2. In the History section, you can view logs of previous queries and interactions. These logs include details such as the source, text, date, and time of each interaction.

Monitor retrieval test queries

Monitoring retrieval test queries helps validate the quality and performance of your Knowledge Base under real-world conditions. By applying monitoring metrics, such as Health Check or Response Latency, you can verify that retrieved chunks are accurate, timely, and aligned with enterprise expectations before moving to production.

You can initiate monitoring directly from the History view in Knowledge Base Retrieval Testing. This reduces navigation steps and preloads query details into the monitoring module for faster validation.

To configure monitoring:

• Select a particular knowledge base and click Retrieval Testing

• Navigate to the History table

• For each test query entry in the table, you can view the Retrieval Result with a Monitor button in the right panel

• Click the 'Monitor' button next to the test query you want to evaluate.

• Go to the Monitor module and configure the monitoring metrics.

A Flow is composed of two essential elements: Triggers and Actions. These elements work together to automate processes, ensuring that workflows are initiated and executed based on defined conditions.

1. Trigger

A trigger is the starting point of a Flow, determining when and how frequently the Flow is executed. It sets the conditions or events that activate the Flow, ensuring it runs at the right time or in response to specific actions.

Types of Triggers:

• Schedule Trigger: Executes the Flow at a specified time or on a recurring schedule, such as daily, weekly, or hourly.

• Webhook Trigger: Activates the Flow in response to an external signal, like receiving an HTTP request or a data submission from an external application.

• Event Trigger: Starts the Flow based on specific events, such as user actions, data updates, or events from external services.

Trigger Configuration:

Triggers can be customized to meet specific requirements, such as:

• Setting time intervals for schedule-based triggers.

• Defining API endpoints for webhook triggers.

• Specifying event parameters for event-based triggers.

2. Action

An action represents a specific task or operation that is executed once the Flow is triggered. Actions define what happens after the trigger event occurs, and they are the building blocks of the workflow. They are responsible for executing the desired operations and achieving the objectives of the Flow.

Types of Actions:

• Data Processing: Manipulates or processes data, such as transforming, filtering, or enriching datasets.

• Communication: Facilitates interaction by sending emails, notifications, or messages to users or systems.

• Integration: Connects with external applications or services to retrieve, update, or send data.

• Decision-m

Key Characteristics of Actions:

• Sequential Execution: Actions are executed in a defined order based on the workflow logic, ensuring that each step is carried out in the correct sequence.

• Flexibility: Multiple actions can be combined to handle complex workflows, enabling more advanced automation.

• Outcome-driven: Each action contributes to a specific objective, such as updating a database, generating a report, or integrating with an API.

By combining well-configured triggers and strategically designed actions, Flows empower users to create complex business logic for developing GenAI-powered apps that automate processes, optimize workflows, and deliver consistent, impactful outcomes.

The Amazon SQS (Simple Queue Service) piece in ZBrain Flow allows you to send messages to AWS SQS queues as part of your automated workflows. It helps you manage reliable, scalable communication between distributed systems without needing to handle complex message broker infrastructure. By adding this connector to your flow, you can queue messages for downstream services or workers to process asynchronously.

This is useful for use cases like task scheduling, batch processing, decoupled microservices, and any workflow that involves passing data between services in a controlled, fault-tolerant way.

How to Use Amazon SQS in ZBrain Flow

Select Amazon SQS (Send Message) as Your Action

1. In your ZBrain Flow, click the + button and search for Amazon SQS.

2. Select Amazon SQS (Send Message) from the list.

Set Up the Amazon SQS Connection

Before using the piece, you need to create a secure connection to your AWS account.

To create the connection:

1. In the ‘Connection’ dropdown, click ‘Create Connection’.

2. In the pop-up window, fill in the following details:

   • Connection Name: A label for this connection (e.g., Amazon SQS)

   • Access Key ID: Your AWS access key

To find the correct region, refer to this document, “”.

Send a Message to an SQS Queue

Once your connection is configured:

1. In the ‘Queue URL’ field, enter the full URL of your target SQS queue.

2. In the ‘Message Body’ field, enter the message you want to send. This can be plain text or JSON, depending on what your consumer service expects.

The information schema is used to create structured data from unstructured data like PDF files, text files etc. This structured information provides insights into important information from large files with the help of LLM models, which can be used for decision-making and querying the bot.

The steps to create and use information schema are as follows -

1. From the available knowledge bases, select one to create an information schema.

2. Navigate to the Information Schema tab and click the ‘+Add New’ button.

1. Click on the ‘+ Add New’ button to open a panel. Here, you can enter a new schema name, select the desired LLM, and choose one of the following options:

   • Fill the column against every page in the file

   • Fill the column against every chunk in the file

   Select your preferred option and click ‘Add’ to proceed.

1. Click the ‘+’ button to add the column details you want to extract from the unstructured data. In the panel, provide a column name, choose the data type (either text or integer), and enter a prompt to retrieve the data. Once you've entered all the details, click Add. If needed, you can add multiple columns by clicking the ‘+’ button for each one.

1. Click the ‘Run Schema’ button at the top right to execute the information schema you have created. Once the run is completed, you can view the results. With the information schema results, you will be able to query data using both the knowledge base and the newly created information schema.

1. Click the ‘Delete Schema’ button to remove the schema you created.

Constant Contact is an online marketing platform that enables businesses to create and manage email campaigns, automate customer engagement, and track performance with analytics. It provides tools for contact management, social media marketing, and e-commerce integration. Businesses use it to streamline communication, nurture leads, and enhance customer relationships.

How to Integrate Constant Contact with ZBrain Flow

To integrate Constant Contact with ZBrain Flow:

1. Click the ‘+’ button in ZBrain Flow.

2. Type "Constant Contact" in the search bar and select it from the available options.

3. Choose an action to begin the setup process.

How to Create a Connection for Constant Contact in ZBrain Flow

To establish a connection between Constant Contact and ZBrain Flow:

1. Enter a Connection Name – Provide a name for easy identification.

2. Click ‘Connect’ – Initiate the authentication process.

3. Log in to Constant Contact – A pop-up window will appear for login credentials.

4. Grant Permissions – Approve the necessary permissions for ZBrain Flow to access Constant Contact.

Once connected, ZBrain Flow can interact with Constant Contact to automate contact management and perform API-based actions.

Available Actions in ZBrain Flow

1. Create or Update Contact

This action allows you to create a new contact or update an existing one in Constant Contact.

Configuration Steps:

• Select an existing connection or create a new one.

• Enter the contact list where the contact should be added.

• Provide the email address of the contact.

• Enter the first and last name of the contact.

2. Custom API Call

This action allows users to perform advanced operations by interacting directly with the Constant Contact API.

Configuration Steps:

1. Select an existing connection or create a new one.

2. Enter the API URL – Use the full URL or a relative path to the base URL.

3. Select the HTTP Method based on your desired action:

   • GET – Retrieve contact or campaign details.

Note: Click ‘f’ to insert dynamic values for flexible configurations.

Contiguity SMS is a messaging service that enables businesses to send SMS notifications, alerts, and updates to customers in real-time. It provides a reliable communication channel with features such as message customization, international number support, and API integration for automated messaging. Businesses use Contiguity SMS for marketing campaigns, customer support, and transactional notifications.

How to Integrate Contiguity with ZBrain Flow

To integrate Contiguity with ZBrain Flow:

1. Click the ‘+’ button in ZBrain Flow.

2. In the search bar, type "Contiguity" and select it from the available options.

3. Choose an action from the list to begin the setup process.

How to Connect Contiguity to ZBrain Flow

To establish a connection between Contiguity and ZBrain Flow:

1. Enter Connection Details

• Connection Name – Provide a name for easy identification.

• API Key – Enter the authentication key for secure access.

2. Obtain the API Key from Contiguity

Follow these steps to retrieve your API key:

1. Log in to your Contiguity account.

2. Navigate to the Dashboard.

3. Go to the API Settings section.

4. Generate a new API key or copy an existing one.

Once you have the API key, use it to authenticate API requests.

3. Save the Connection

Click ‘Save’ to finalize the connection.

Available Actions in ZBrain Flow

1. Send SMS

This action allows users to send SMS messages directly through Contiguity.

Configuration Steps:

• Select an existing connection or create a new one.

• Enter the recipient's phone number – Use the international format without spaces (e.g., +12065551234).

• Enter the message content – Provide the text to be sent via SMS.

2. Custom API Call

This action enables users to execute API requests, such as retrieving, creating, updating, or deleting records.

Configuration Steps:

• Select an existing connection or create a new one.

• Enter the URL – Specify the API endpoint required for the request.

• Choose the HTTP Method that matches your request type:

  • GET – Retrieve records or details.

The Azure Translation piece in ZBrain Flow provides integration with Microsoft's AI-powered translation service, allowing you to convert text and documents between languages directly from your workflows. This connector enables you to automate language translation without manual intervention. With Azure Translation integration, you can build automations that process multilingual content, localize materials, enable cross-language communication, and break down language barriers in your business processes. This piece is ideal for organizations working in multiple regions, handling international communications, or needing to process content in various languages as part of their automated workflows.

How to Use Azure Translation in ZBrain Flow?

Select Azure Translation as Your Connection

• Click on the '+' button in the Flow and search for Azure Translation.

• Select Azure Translation.

• Decide on the action you need, then select it. ZBrain Flow provides two options:

  • Translate Text – Convert text content from one language to another.

How to Connect to Azure Translation Services

Before using any Azure Translation actions in ZBrain Flow, you'll need to set up a connection to your Azure Translator resource. This is a one-time setup that will allow you to access language translation services securely.

To create your Azure Translation connection:

• From any Azure Translation action, click on the connection dropdown and select 'Create connection'.

• In the popup window that appears, you'll need to:

  • Enter a descriptive 'Connection Name' to identify this Azure connection

  • In the 'Translator Endpoint' field, enter the endpoint URL for your Azure Translator resource

To obtain your Azure Translator credentials:

1. Log in to the Azure Portal

2. Navigate to your Azure Translator resource

3. Go to the "Keys and Endpoint" section in the left sidebar

4. Copy the endpoint URL and one of the access keys (Key 1 or Key 2)

How to Translate Text Between Languages

Configuration Steps:

• Connect to your Azure Translation Services following the steps in the "How to Connect to Azure Translation Services" section.

• In the 'Text' field, enter the content you want to translate. This can be a sentence, paragraph, or longer text.

• In the 'Target Language' field, enter the language code for the desired output language. Examples: 'es' for Spanish, 'fr' for French, 'de' for German, 'ja' for Japanese.

• In the 'Source Language' field, optionally specify the language code of your input text. If left empty, Azure will automatically detect the source language. Specifying the source language can improve accuracy in some cases.

How to Translate Documents in Different Languages

Configuration Steps:

• Connect to your Azure Translation Services following the steps in the "How to Connect to Azure Translation Services" section.

• In the 'Document URL' field, enter the web address where your document is accessible. This must be a publicly accessible URL that Azure can download from.

• In the 'Target Language' field, enter the language code for the desired output language.

• In the 'Source Language' field, optionally specify the language code of your document.

Crypto functions in ZBrain Flow allow users to perform essential cryptographic operations, including hashing, generating HMAC signatures, and creating strong passwords. These actions help ensure data security, authentication, and encryption when processing sensitive information.

How to Integrate Crypto with ZBrain Flow?

To use Crypto in ZBrain Flow, follow these steps:

1. Click the '+' button in the Flow and search for Crypto.

2. Select the specific Crypto action you want to configure.

3. Choose from the following available actions:

   • Text to Hash: Convert text into a hash value using various hashing algorithms.

   • Generate HMAC Signature: Encrypt text using a secret key and hashing algorithms.

   • Generate Password: Create a secure, random password with customizable length and character sets.

Text to Hash

The Text to Hash action converts a given text string into a hashed value using cryptographic hash functions. Hashing is commonly used for password storage, data integrity verification, and digital signatures.

How to Configure Text to Hash in ZBrain Flow?

1. Select the hashing algorithm from the available options:

   • MD5 – A widely used cryptographic hash function that generates a 128-bit hash.

   • SHA256 – A more secure hashing function generating a 256-bit hash, commonly used in blockchain and encryption.

   • SHA512 – A stronger cryptographic function that produces a 512-bit hash value.

Generate HMAC Signature

The Generate HMAC Signature action creates a cryptographic hash-based message authentication code (HMAC). This is used to verify the integrity and authenticity of a message using a secret key. HMAC is commonly used in API authentication, digital signatures, and secure data transmission.

How to Configure Generate HMAC Signature in ZBrain Flow?

1. Enter the secret key in the Secret Key field. This key is used to sign the message securely.

2. Select the secret key encoding format from the available options:

   • UTF-8 – Standard encoding for text data.

   • Hex – Represents the key in hexadecimal format.

Generate Password

The Generate Password action creates a strong, random password with a defined length and character set. This is useful for enhancing security by generating unpredictable passwords for authentication, encryption, and system access.

How to Configure Generate Password in ZBrain Flow?

1. Enter the desired password length (Maximum up to 256 characters).

2. Choose the character set for the password generation:

   • Alphanumeric – Generates a password containing letters and numbers (A-Z, a-z, 0-9).

   • Alphanumeric + Symbols – Generates a more complex password including special characters (e.g., !@#$%^&*).

The Amazon SNS (Simple Notification Service) piece in ZBrain Flow allows you to send messages directly to SNS topics as part of your automated workflows. This integration enables seamless communication between your business systems and any subscribed services or endpoints, including email, SMS, mobile push, and other AWS services. By incorporating this piece into a flow, you can trigger real-time notifications or updates based on workflow events, without writing custom backend code.

This connector is useful for scenarios such as system alerts, status updates, user notifications, and any workflow that requires instant broadcasting of messages to distributed services or users via AWS SNS.

How to Use Amazon SNS in ZBrain Flow

Select Amazon SNS (Send Message) as Your Action

1. In your ZBrain Flow, click the + button and search for Amazon SNS.

2. Select Amazon SNS (Send Message) from the list of available actions.

Set Up the Amazon SNS Connection

Before using Amazon SNS, you need to configure a connection with your AWS account. This setup is required once and can be reused across flows.

To create a new connection:

1. In the Connection dropdown, click ‘Create Connection’.

2. In the pop-up window that appears, fill in the following fields:

   • Connection Name: A label to identify this connection (e.g., Amazon SNS)

   • Access Key ID: Your AWS access key with SNS permissions

Note: You can find your credentials in the AWS Management Console under IAM > Users > Security credentials. Refer to this document to find the correct region. -

How to Send a Message to an SNS Topic

Once your connection is configured:

1. In the ‘Topic’ field, select the SNS topic you want to send the message to.

2. In the Message field, enter the content you want to send. This can be plain text or structured data (e.g., JSON) depending on how your subscribers are set up to receive messages.

Approval in ZBrain Flow is a workflow tool for managing approvals and permissions. It allows users to create approval requests, generate approval links, and pause flows until a decision is made, ensuring seamless process control and authorization handling.

How to Integrate Approval with ZBrain Flow

1. Click the “+” button in the Flow: Open your ZBrain Flow and click the plus sign to add a new step.

2. Search for “Approval”: Type “Approval” in the search bar to see the available approval-related actions.

3. Choose the Desired Task: Select the approval function that fits your workflow requirements.

Tasks You Can Perform with Approval

• Wait for Approval: Pauses the workflow and waits for a user to approve or reject a request before proceeding.

• Create Approval Links: Generates approval links that can be shared externally without pausing the flow, allowing for asynchronous decision-making.

Wait for Approval

Pauses the workflow and waits for a user to approve or reject a request before proceeding.

How to Configure Wait for Approval in ZBrain Flow?

1. Add the Step

   • In your ZBrain Flow, click the + button to add a new step.

   • Search for “Approval” and select Wait for Approval from the available options.

2. Place in the Workflow

Create Approval Links

Generates approval links that can be shared externally without pausing the flow, allowing for asynchronous decision-making.

How to Configure Create Approval Links in ZBrain Flow?

1. Add the Step

   • In your ZBrain Flow, click the + button to add a new step.

   • Search for “Approval” and select Create Approval Links from the available options.

2. Place in the Workflow

The Bubble piece in ZBrain Flow enables seamless integration with Bubble applications, allowing you to automate data management and interact with your Bubble database effortlessly. With this connector, you can create, update, delete, and retrieve data objects (Things) directly within your workflows. This integration simplifies communication with your Bubble backend, eliminating the need for custom API setups.

How to Use Bubble in ZBrain Flow

1. Click on the ‘+’ button in the Flow, search for ‘Bubble,’ and select it.

2. Choose the required action from the available options:

   • ‘Create Thing’ – Add a new data object.

   • ‘Delete Thing’ – Remove an existing thing.

   • ‘Update Thing’ – Modify an existing thing.

   • ‘Get Thing’ – Retrieve a specific thing by its ID.

How to Create a Thing in Bubble

1. In the ‘Typename’ field, enter the name of the Bubble data type you want to create.

2. In the ‘Body’ section, provide the key-value pairs for the fields you want to populate.

How to Delete a Thing in Bubble

1. In the ‘Typename’ field, enter the name of the Bubble data type you want to delete.

2. In the ‘Thing ID’ field, provide the unique ID of the item to be deleted.

How to Update a Thing in Bubble

1. In the ‘Typename’ field, enter the name of the Bubble data type you want to update.

2. In the ‘Thing ID’ field, provide the unique ID of the item to be updated.

3. In the ‘Body’ section, define the fields and values you want to update.

How to Retrieve a Thing in Bubble

1. In the ‘Typename’ field, enter the name of the Bubble data type you want to retrieve.

2. In the ‘Thing ID’ field, provide the unique ID of the item you want to fetch.

How to List Things in Bubble

1. In the ‘Typename’ field, enter the name of the Bubble data type you want to retrieve a list from.

2. (Optional) Add constraints:

   • In the ‘Constraint’ field, define filtering conditions.

   • In the ‘Field’ section, specify the data field to apply the constraint.

The Brilliant Directories piece in ZBrain Flow enables seamless integration with your Brilliant Directories site, allowing you to automate user management and interact with the platform via API calls. With this connector, you can create new users, manage memberships, and execute custom API calls, streamlining directory-based workflows. Whether you're handling user registrations or performing advanced API operations, Brilliant Directories integration helps automate these tasks efficiently.

How to Use Brilliant Directories in ZBrain Flow?

Step 1: Select Brilliant Directories as Your Connection

1. Click on the ‘+’ button in the Flow and search for ‘Brilliant Directories’.

2. Select Brilliant Directories.

3. Choose the required action from the available options:

   • Create New User – Add a new user to your Brilliant Directories site.

   • Custom API Call – Make a request to a specific API endpoint for custom operations.

Create a New User in Brilliant Directories

Step 1: Create a Connection

1. In the 'Connection' dropdown, click on 'create connection'.

2. A window will appear, prompting you to enter connection details:

   • Provide a unique name in the 'Connection Name' field.

   • In the 'API Key' field, enter your API key which can be generated from here:

• Click 'Save' to establish the connection.

Step 2: Configure User Details

1. Enter the user’s email in the 'Email' field.

2. Set a password for the new user account in the 'Password' field.

3. Provide the subscription ID from your website in the 'Subscription ID' field.

4. If needed, add extra user details in the 'Meta' field.

Make a Custom API Call in Brilliant Directories

Step 1: Create a Connection

• Follow the instructions in the ‘Create a Connection’ section from ‘Create a New User in Brilliant Directories’ to set up the connection.

Step 2: Configure API Request

1. Select the HTTP method from the 'Method' dropdown.

2. If needed, add headers by clicking 'Add Item' in the 'Headers' section (authorization headers are automatically injected from the connection).

3. Add query parameters in the 'Query Parameters' field by clicking 'Add Item.'

4. Enter the request body in the 'Body' field if required.

Step 3: Set Additional Options

1. Enable 'No Error on Failure' if you want the flow to continue even if the request fails.

2. Set the request timeout in the 'Timeout (in seconds)' field.

You can update the existing knowledge base created with the knowledge graph as an RAG definition by adding a chunk to it.

• Select the existing knowledge base created using a knowledge graph.

• Click on the knowledge base file name.

The Azure Cosmos DB piece in ZBrain Flow provides integration with Microsoft's globally distributed, multi-model database service, allowing you to interact with your NoSQL data directly from your workflows. This powerful connector enables you to create, read, update, and delete documents without manual intervention. With Azure Cosmos DB integration, you can build automations that store application data, retrieve information based on queries, maintain document collections, and synchronize information between your database and other business systems. This piece is ideal for organizations looking to incorporate cloud-based NoSQL database operations into their automated processes, enabling scalable, high-performance data storage and retrieval for a wide range of applications.

How to Use Azure Cosmos DB in ZBrain Flow

The Azure Document Intelligence piece in ZBrain Flow provides integration with Microsoft's AI-powered document processing service, allowing you to extract information from various document types directly from your workflows. This powerful connector enables you to automate document analysis without manual intervention. With Azure Document Intelligence integration, you can build automations that extract structured data from documents, process invoices and receipts, analyze tables and forms, and convert unstructured documents into actionable data. This piece is ideal for organizations looking to streamline document processing, automate data entry, reduce manual handling of documents, and incorporate intelligent document extraction into their business processes.

How to Use Azure Document Intelligence in ZBrain Flow

Box is a cloud-based content management and file-sharing platform that enables businesses to securely store, access, and collaborate on files from anywhere.

How to Integrate Box with ZBrain Flow

To integrate Box with ZBrain Flow:

1. Click the ‘+’ button in ZBrain Flow.

The Calendly piece in ZBrain Flow enables seamless scheduling automation by integrating with your Calendly account. This component allows you to send custom API requests to Calendly, enabling operations such as retrieving events, scheduling meetings, and managing availability. With this integration, you can automate scheduling tasks directly within your workflows.

How to Use Calendly in ZBrain Flow

Step 1: Select Calendly as Your Connection

Click on the '+' button in the Flow and search for Calendly. Select Calendly.

Step 2: Create a Connection

The Clearout piece in ZBrain Flow provides integration with Clearout's email verification platform, allowing you to validate email addresses directly from your workflows. This connector enables you to instantly check email deliverability, reduce bounce rates, and ensure email list quality without manual verification. With Clearout integration, you can build automation that verifies email addresses at critical points in your business processes, cleans contact lists, and maintains high-quality email databases. This piece is ideal for businesses looking to improve email deliverability, protect their sender reputation, and ensure they're communicating with valid contacts.

How to Use Clearout in ZBrain Flow?

Step 1: Select Clearout as Your Connection

Gmail is a robust email service provided by Google. It enables users to send, receive, and manage emails efficiently with powerful spam filtering, advanced search capabilities, and seamless integration with other Google services, making it ideal for personal and professional communication and collaboration.

To integrate Gmail with ZBrain Flow, follow these detailed steps:

• Click the '+' button in the Flow and search for Gmail in the ZBrain Flow.

• Select the specific task from the list you want to include in the Flow. You can perform the below listed tasks:

Connections in ZBrain Flow facilitate seamless integration with external platforms, enabling users to securely retrieve and manage connection details. This feature is particularly beneficial when connection external IDs are dynamically received, such as through a webhook message. Instead of manually specifying a connection, this action allows you to dynamically fetch the appropriate connection based on the received external ID, ensuring greater flexibility and automation in your workflows.

Finding the Connection External ID

The external ID can be found on the connection settings page. Hover over the connection name to reveal and copy the external ID. This helps retrieve connection values using their external IDs from the project.

Facebook Pages are public profiles designed for businesses, brands, organizations, and public figures to connect with their audience. Unlike personal profiles, Pages allow for content publishing, audience engagement, and advertising, making them essential for marketing and community building. With features like posts, events, insights, and messaging, Facebook Pages help businesses establish an online presence and interact with followers effectively.

How to Integrate Facebook Pages with ZBrain Flow

1. Click the ‘+’ button in ZBrain Flow.

The Apify piece in ZBrain Flow provides integration with Apify's web scraping and automation platform, allowing you to extract data from websites and run web automation tasks directly from your workflows. This powerful connector enables you to leverage Apify's extensive library of web scrapers (called Actors) without manual intervention. With Apify integration, you can build automations that gather web data, run scheduled web scraping jobs, process extracted information, and incorporate web data into your business processes. This piece is ideal for organizations looking to automate data collection from websites, monitor online content, extract structured information from the web, and integrate web data into their digital ecosystem.

How to use Apify in ZBrain flow?

The Delay piece in ZBrain Flow allows you to introduce timed pauses in your workflow execution. This essential component helps control the timing and pacing of your automation processes, whether you need to wait for a specific duration or until a particular date and time. By strategically incorporating delays, you can schedule actions, prevent API rate limiting, allow time for external processes to complete, or ensure your workflow executes at the optimal moment.

How to Use Delay Piece in ZBrain Flow?

Step 1: Select Delay as Your Connection

The Files Helper piece in ZBrain Flow provides essential file management capabilities for your workflows. This versatile component allows you to read, create, and manipulate files within your automations. With Files Helper, you can process file contents, generate new files with custom data, convert file encodings, and verify file types - all crucial operations when building workflows that involve document processing, data transformation, or content generation.

How to Use Files Helper in ZBrain Flow?

• Click on the '+' button in the Flow and search for Files Helper.

GenerateBanners is an API-driven platform that allows users to create high-quality banners, images, and PDFs dynamically. It automates the design process by enabling users to customize templates, generate visuals programmatically, and streamline content creation. Ideal for marketing teams, e-commerce platforms, and branding purposes, GenerateBanners simplifies large-scale banner generation without manual design effort.

How to Integrate GenerateBanners with ZBrain Flow

To seamlessly integrate GenerateBanners with ZBrain Flow, follow these steps:

Google My Business is a free tool from Google that allows businesses to manage their online presence across Google Search and Maps. It enables businesses to create and update profiles, engage with customers through reviews and messages, showcase products and services, and gain insights into customer interactions. This tool helps businesses improve their local visibility and connect with potential customers more effectively.

How to Integrate Google My Business with ZBrain Flow

To integrate Google My Business with ZBrain Flow, follow these steps:

Flowise is an open-source, low-code tool designed to help developers build customized applications using Large Language Models (LLMs). It provides a drag-and-drop interface to create and manage LLM orchestration flows and AI agents.

How to Integrate Flowise with ZBrain Flow

To integrate Flowise, please follow these steps:

1. Click the '+' button in ZBrain Flow and search for Flowise.

The Code piece in ZBrain Flow provides powerful scripting capabilities that allow you to execute custom code within your workflows. This component enables advanced data manipulation, complex calculations, custom algorithms, and more services that may not be covered by standard components. With the Code piece, you can write custom logic in either JavaScript/TypeScript or Python, depending on your preference and requirements.

How to Use the Code Piece in ZBrain Flow?

Step 1: Select Code as Your Connection

The Heartbeat component is an all-in-one platform for community businesses that helps monitor and alert.

How to Connect Heartbeat Component to ZBrain Flow?

To integrate Heartbeat with ZBrain Flow, follow these steps to select and utilize the desired capabilities:

1. Search for the Heartbeat Component

Frame is a cloud-based virtual desktop and application delivery platform that allows businesses to securely access and manage applications from anywhere. It provides a scalable, high-performance solution for remote work, enabling seamless application streaming and virtual desktop infrastructure (VDI).

How to Integrate Frame with ZBrain Flow

To integrate Frame with ZBrain Flow:

Formbricks is an open-source form and survey platform that allows businesses to collect, analyze, and manage user feedback efficiently. It offers powerful customization options, seamless integrations, and real-time analytics to improve data-driven decision-making.

How to Integrate Formbricks with ZBrain Flow

To integrate Formbricks with ZBrain Flow:

What is a Flow in ZBrain?

A Flow in ZBrain Builder is an operational blueprint that visually defines how your agent, application, or business workflow interacts with data, applies logic, makes decisions, and executes actions. Using the intuitive, low-code Flow module, you can quickly develop AI-powered apps and agents for a wide range of use cases with ease by designing sophisticated workflows that integrate AI models, knowledge bases, logical decision points, helper methods, and third-party tools.

Flows can automate various scenarios, including compliance checks, sending notifications and generating reports. By visually structuring these processes, you can standardize internal processes, automate repetitive tasks, improve collaboration, and ensure seamless communication throughout your organization.

Key features of ZBrain Flow

Below are the key features of ZBrain Flow that allow you to build and customize workflows for your agents.

1. Low-code interface

Effortlessly build complex workflows with a visual, low-code interface that allows for easy configuration and customization of components.

1. GenAI capabilities

Harness cutting-edge AI models such as GPT-4, Mistral, and Llama 3 to automate processes, enhance efficiency, and support intelligent decision-making.

1. Component connectivity

Integrate AI models, knowledge bases, and external tools to create a powerful AI agent for diverse applications.

1. Robust logic elements

Incorporate advanced logic elements like conditional branching, time delays, and date-format conversions to build highly customized workflows.

1. Custom scripting

Leverage scripting capabilities that enable personalized functionality, including configurable API keys and AI-generated code for dynamic and unique workflows.

Dive deeper into ZBrain Flow:

Asana in ZBrain Flow is a project management platform designed to help teams organize, track, and manage their work. With Asana, you can create tasks, monitor progress, and collaborate seamlessly, ensuring that your team stays aligned on goals and deadlines.

How to Integrate Asana with ZBrain Flow?

1. Click the “+” Button in the Flow Open your ZBrain Flow and select the plus sign to add a new step.

2. Search for “Asana” Type “Asana” in the search bar to view the available tasks.

3. Choose the Desired Task Select the specific Asana task you want to include in your Flow.

Tasks You Can Perform on Asana

• Create Task: Add a new task to your Asana project by specifying details such as task name, description, assignee, and due date.

• Custom API Call: Make a tailored API request to Asana’s endpoints, providing you with the flexibility to access or modify Asana data beyond the standard actions.

Create Task

Add a new task to your Asana project by specifying details such as task name, description, assignee, and due date.

How to Configure the "Create Task" Action in Asana Using ZBrain Flow?

Step 1: Add the “Create Task”

1. Open Your Flow

   • From your ZBrain Flow, locate where you want to create a new Asana task.

2. Click the “+” Button

   • Click the plus sign (+) to add a new step in your flow.

Step 2: Connect Your Asana Account

1. Select a Connection

   • In the Connection dropdown, choose the Asana account you want to use.

Step 3: Configure the Task Details

1. Workspace

   • Select the Asana workspace in which you want to create the task.

   • A workspace is where your projects and tasks are organized in Asana.

2. Project

Step 4: Save and Test

1. Save the Step

   • Once you’ve entered all the necessary details, click Save or Done to confirm your settings.

2. Run or Test the Flow

Custom API Call

Make a tailored API request to Asana’s endpoints, providing you with the flexibility to access or modify Asana data beyond the standard actions.

How to Configure Custom API Call in Asana Using ZBrain Flow?

1. Add the Custom API Call Step

   • In your ZBrain Flow, click the “+” button to add a new step.

   • Type “Asana” in the search bar and select the Custom API Call action from the available options.

Amazon S3 (Simple Storage Service) is a scalable object storage service provided by Amazon Web Services (AWS). It allows users to store and retrieve any amount of data at any time, making it ideal for backup, archiving, big data analytics, content distribution, and cloud applications.

How to Connect Amazon S3 to ZBrain Flow ?

To integrate Amazon S3 with ZBrain Flow, follow these detailed steps:

Step 1: Select Amazon S3 as Your Connection

• Click on Amazon S3 in ZBrain Flow.

• Choose between two available actions.

Step 2: Configure the Amazon S3 Connection

To establish a connection between Amazon S3 and ZBrain Flow, enter the following details:

1. Connection Name

   • Provide a unique and recognizable name for the connection.

2. Access Key ID

   • This is required to authenticate with Amazon S3.

Triggers Available in ZBrain Flow

• New File Upload: This trigger is activated whenever a new file is uploaded to the specified Amazon S3 bucket.

• Specify the folder path within the S3 bucket where your files are stored (e.g., /root/subfolder/).

Actions Available in ZBrain Flow

1. Upload File – Uploads a file to an Amazon S3 bucket.

   • Enter the file: Specify the file you want to upload.

   • Provide a file name (without extension): Ensure the file name does not include an extension.

   • Select the ACL (Access Control List)

By following these steps, you can successfully integrate Amazon S3 with ZBrain Flow, enabling seamless file management and data retrieval for your workflows.

This section outlines how users can review and manage their file content within the knowledge base. The Knowledge Source feature allows users to edit metadata, view technical details such as chunking strategy and embedding time, and manage document chunks by editing, disabling, or adding new ones. Users can also generate summaries of their files using available models, with options to download or delete the content as needed.

How to review file content?

• From the ‘Knowledge’, click on any knowledge base you created.

• You will be directed to the ‘Knowledge Source’ tab. There, click on the file name again.

• You can view the total number of paragraphs divided into chunks. In addition, you will see the character count of each chunk, along with its ID and retrieval count.

• You can view additional details of the data, including technical parameters such as file name, size, update and modification date details, source, chunking strategy, total number of chunks, average paragraph length, and embedding time.

Updating metadata

• Select the appropriate option from the ‘Document Type’ dropdown, edit the metadata, and click ‘Save’ on the right-side panel titled ‘Document Metadata’

• Once you save the updated metadata, it will be applied to all chunks. Click on any chunk to view the updated metadata.

How to check the semantic relationships of the entities in a knowledge graph?

When you create a Knowledge Base with the RAG definition set to Knowledge Graph, ZBrain automatically processes every document, extracts entities and the semantic relationships between them, and saves them as a property graph, a network of nodes (the extracted entities) linked by edges (the relationships between those entities). To explore that graph:

a. From the ‘Knowledge’, click on any knowledge base you created using the knowledge graph option.

b. You will be directed to the ‘Knowledge Source’ tab. There, click on the file name to view the file content and its associated knowledge graph.

c. Click Knowledge Graph on the left-hand side bar.

The centre panel renders an interactive force-directed graph:

• Nodes (coloured circles): Individual entities, people, products, APIs, components, dates, etc. You can hover to see the canonical name, type, source sentence and chunk ID.

• Edges (grey lines): Represent which entities (nodes) are linked.

How to update an existing knowledge base?

You can update an existing knowledge base created with either vector store or the knowledge graph option by adding/deleting chunks.

• To add a new chunk, click the ‘+Add Chunk’ button. Enter the content in the provided text box, then click the ‘Save’ button to finalize.

• The chunk is saved immediately and becomes part of the existing knowledge base.

• You can also update the existing chunk. Click the ‘pencil' icon at the rightmost corner and edit the chunk, and click 'Save’.

• Once you edit or add a chunk, you can track the status in the dashboard.

• You can also delete a chunk from the knowledge base. Click on the delete button at the right end of the chunk and confirm.

• Once confirmed, the chunk will be deleted from the knowledge base.

• Click on the three dots next to ‘+Add Chunk’ to access three options:

  • Download File: Click this option to download the file.

  • Review File

• If the knowledge base was created using the Knowledge Graph option, any newly added or edited chunk will automatically update the graph to include the new entities and their relationships. You can find the updated knowledge graph, which now includes the new entity.