Tool Management

  • Updated on Oct 31, 2025
  • Published on Oct 31, 2025
  • 2 minute(s) read
  • Suara Atay
 Prev Next

Tool Management

Knovvu Virtual Agent provides a comprehensive tool management system that enables agents to extend their capabilities beyond conversation by interacting with external systems and executing complex business logic. The platform supports three types of tool integrations: API Integration for direct system connections, Workflow for custom business process automation and MCP tools for standardized, context-aware access to distributed resources and services through the Model Context Protocol.

Tool Types

1. API Integration

API Integration tools provide a streamlined method for connecting Knovvu Virtual Agent to external REST APIs and web services. These tools enable agents to retrieve real-time data, trigger actions in third-party systems, and synchronize information across platforms without requiring complex coding.

The API Integration interface simplifies the connection process by providing structured forms for endpoint configuration, authentication setup, and parameter mapping. Once configured, these tools become immediately available to agents, who can intelligently determine when to invoke them based on conversation context and user needs.

image.png

2. Workflow

Workflow tools empower users to design custom automation flows using a visual, drag-and-drop interface. These tools excel at handling multi-step processes, conditional logic, and complex orchestrations that involve multiple systems or require business-specific rules.

Unlike simple API calls, Workflows can maintain state across multiple steps and implement branching logic based on conditions. This makes them ideal for complex business processes like order fulfillment, claim processing, or multi-system data validation.

image.png

3. Model Context Protocol (MCP)

Model Context Protocol is a standardized interface specification that defines how AI models communicate with external systems to access context, execute actions, and retrieve information. For more detailed information, please visit here.

Tool Configuration

Every tool in Knovvu Virtual Agent requires foundational configuration that helps agents understand when and how to use it effectively.

image.png

  • Tool Name: A unique identifier for the tool that agents reference during execution. Best practices suggest using clear, descriptive names that indicate the tool's purpose (e.g., "check_inventory_status" rather than "api_call_1").

  • Tool Description: A comprehensive explanation of the tool's functionality, written in natural language that helps the LLM understand the tool's purpose, capabilities, and appropriate use cases. This description is crucial for agent decision-making.

  • Tool Parameters: Parameters transform static tool definitions into dynamic, context-aware operations. For example, instead of having separate tools for each customer lookup, a single "get_customer_details" tool with an "email" parameter can handle all customer queries by accepting different email addresses extracted from the conversation.
    A tool parameter has the following elements:

    • Parameter Name: The technical identifier used in API calls or workflow logic.
    • Parameter Type: Defines the data type expected for the parameter. Knovvu VA supports the following types: string, number, boolean, object, array, integer.
    • Parameter Description: Detailed explanation of what the parameter represents.
    • Predefined Options (Enum): It is used when a parameter’s possible values are limited to a fixed set of choices. For example, a tool parameter named language might have predefined options like English, Turkish and Arabic.
    • Required Parameter: Toggle switch that indicates whether the parameter must be provided for successful tool execution.

image.png