This article explains how administrators add, configure, and manage Tools for a Copilot in Yeeflow.
Tools enable Copilot to go beyond conversation and perform real system actions, such as querying data, creating or updating records, and starting business workflows.
What Are Tools in Copilot
Tools define the actions Copilot is allowed to perform and the scope of those actions.
With Tools configured, Copilot can:
Query real-time data in applications
Create, update, or delete records
Start approval or business workflows
Call other AI Agents to perform sub-tasks
Return execution results to users or use them in follow-up steps
If no tools are configured, Copilot can only provide informational responses and cannot perform actions.
The 6 Tool Types Supported by Copilot
Copilot currently supports the following six core tool types.
Retrieves real-time records from a selected data list, with support for filters, selected fields, and sorting.
Typical use cases:
Query inventory quantities
Find tickets submitted by a specific user
Filter approval requests with status “Pending”
Automatically creates new records in a data list based on user instructions, and intelligently fills field values.
Typical use cases:
Create a ticket from a user description
Register a new product or supplier
Create inventory inbound or issue records
Updates specific fields of a record based on its ID.
Typical use cases:
Mark an order as “Completed”
Adjust inventory quantities
Update owner or remarks
Permanently deletes a record based on its ID.
Typical use cases:
Remove discontinued products
Delete incorrect or test data
⚠️ Deletion actions are irreversible and should be enabled with caution.
Automatically triggers an approval or business workflow and fills in workflow fields based on user input.
Typical use cases:
Start a purchase request
Submit a travel approval
Initiate an expense reimbursement process
Calls another AI Agent to complete a sub-task, enabling multi-agent collaboration.
Typical use cases:
An inventory Copilot calls a replenishment analysis agent
A contract Copilot calls a risk evaluation agent
Adding a Tool to a Copilot
Before Copilot can use a tool, the tool must be added to the Copilot.
Steps to Add a Tool
Go to Application Management
Open AI Agent → Copilot Management
Select a Copilot
Switch to the Tools tab
Click Add Tool
Select a tool type
Configure the tool settings
Save the configuration
After saving, the tool appears in the Copilot Tools list.
Common Tool Configuration (Details)
Regardless of tool type, every tool shares a set of common configuration options.
1. Name & Description
Defines the tool’s name and description to help Copilot understand:
What the tool does
When it should be used
Clear naming and descriptions improve Copilot’s ability to select and execute the correct tool.
2. Application & Data Source
When creating a tool, you must specify:
The Application it belongs to
The Data Source it operates on (such as a data list, document library, or workflow form)
You can click Open next to the application or data source to view it in a new window.
3. Credentials to Use
Controls which identity is used when the tool runs:
End user credentials
Executes the tool using the current end user’s identitySpecific user credentials
Executes the tool as a specific userDefaults to the current user
If the selected user is invalid or lacks permission, execution will fail with an error
This setting defines the permission boundary of the tool.
Tool Input Mapping
Each tool supports input mapping to pass values from conversations, variables, or constants into the tool.
Common Input Fields
Input Name – Logical name of the input
Input Type – Data type (text, number, attachment, etc.)
Fill Using – How the value is populated
Value – The actual value or expression
Dynamically Fill with AI
Values are dynamically extracted by AI from the user’s natural language input
Suitable when users describe information directly in conversation
Reduces manual configuration and improves natural interaction
Example:
User says: “I want to apply for a business trip starting on March 15.”
Copilot automatically extracts the start date and fills the corresponding field.
Custom Value
Uses a fixed or predefined value
Does not depend on user input
Suitable for fields with stable or rule-based values
Example:
Setting a default status, request type, or constant parameter.
📖 For detailed usage and differences between Dynamically fill with AI and Custom value, see: Usage and Differences Between “Dynamically fill with AI” and “Custom value” in AI Agent Tool Inputs
Completion Settings (Tool Execution Behavior)
Defines how Copilot behaves after a tool is triggered.
Waiting for Response
Copilot waits for the tool to finish execution
Receives outputs and continues processing
Recommended when results are needed immediately.
Do Not Respond
Copilot triggers the tool and continues without waiting
Suitable for asynchronous background tasks
Tool Outputs
After a tool runs, it may return:
Workflow ID
Execution status
Record ID
Error messages
Auto-generated fields
Processing results
These outputs can be used in follow-up logic or returned directly to the user.
Tool Configuration Best Practices
Enable only the tools required for the Copilot’s role
Use Instructions to clearly define when tools should be used
Configure Update and Delete tools cautiously
Use Test Run to validate tool behavior before deployment