Navigate a digital worker¶

Introduction¶

The Agentic Platform has a library of digital workers. This guide will introduce you to the various components of a digital worker and their functions.

Access digital workers¶

1. Log in to your Agentic Platform workspace
2. Ensure you are in the Agentic Studio by navigating to the Utility Menu (3 icons near the top right)
3. Click the Studio Selector (four squares)
4. Choose Agentic Studio
5. Select AI on the purple navigation menu (far left)
6. Here you can view a library of the Digital Workers

View Details¶

1. Click the title to select a Digital Worker
2. There are two tabs: Overview and Graph
   1. Overview: provides a broad view of of the Digital Worker’s metrics such as number of interactions and a graph showing interactions vs abandonments
   2. Graph: provides visualization of all the tools and skills (sub agents) that are part of the Digital Worker
3. On the right sidebar, select Compose (pencil icon) to access further details
4. Instructions will display by default, these describe the work performed by the Digital Worker
5. While not required, we recommend as a best practice the following format for your instructions:
   1. Brief Description
   2. Goal
   3. Rules
   4. Steps (what a human would also do)
   5. Validation
6. While the next section will cover tools, it important to note there are no tools specified in the instructions, the Digital Worker will determine what tools are used and when

Tools and Skills¶

Digital workers leverage both tools and sub-agents to accomplish their work. Tools operate with fixed input and output parameters, providing deterministic and predictable results for specific functions. In contrast, sub-agents are capable of independent decision-making and can adapt their behavior based on feedback and learning from previous interactions. This combination allows digital workers to handle both routine, structured tasks through reliable tools while managing complex, dynamic situations through intelligent sub-agents.

Tools¶

1. Navigate to the Tools tab to view assigned tools & settings
2. Each tool contains an opportunity to select whether you want a human-in-the-loop to approve the decision, this can be selected using the Approval mode:
   1. If required:
      1. Approval type: internal confirmation or external authorization
      2. Prompt: helps agent make the decision on invoking the tool
      3. Tag: only if successful & also optional
      4. Output prompt: used to extract necessary information from the tool outputs
      5. Step name: name used by the Digital Worker to refer to the tool
   2. If no limit (not required):
      1. The workflow will run as designed with no human interaction

Skills¶

Additional Skills can be added to your Digital Worker, enhancing their capabilities. These enhanced capabilities can enable your Digital Worker to perform a wider range of tasks and provide greater assistance.

1. Navigate to Skills
2. Click the skill to view the prompt & sub agent attributes
3. The Prompt field can be used to aid the Digital Worker when deciding which Skills to handoff to
4. Skills can also contain collections of tools

Triggers¶

Triggers add actionability to a Digital Worker. Once a trigger has been added to a Digital Worker, it will observe and take action when the context is deemed to be relevant.

1. Navigate to Triggers
2. Depending on what connectors you have added to the system, any events coming from the connector side can act as a trigger
3. Multiple triggers can initiate the agent

Settings¶

Here you can access and customize key aspects of the Digital Worker’s behavior. Each setting allows you to tailor how the Digital Worker executes tasks, collaborates with other Digital Workers and interacts with humans when needed.

1. Human-in-the-Loop (HITL): Configure escalation rules for scenarios where the system requires human review or encounters an error
   1. Supported notification channels:
      1. Slack
      2. Microsoft Teams
      3. Email
      4. Conditional routing (logic-based HITL flows)
2. Default Prompt: Defines how other agents should interact with this agent when using it as a sub-agent
3. Skill Attributes: Specify the input variables the agent expects to receive when it is invoked by another agent
4. Sub Agent Output Attributes: Define the output variables that the agent will return after completing its task
5. Enable Reflection: Controls whether the agent performs self-reflection at each step of its plan
   1. Off: Default
   2. On: When enabled, the agent pauses to evaluate it's progress, improving reliability at the cost of additional processing time
6. Enable Concurrent Processing: Allows the agent to handle multiple tasks in parallel, increasing throughput for high-volume workflows
7. Agent Type: Select the category to toggle between reasoning LLMS and non-reasoning LLMS. You can also set values for the levels of reasoning and verbosity

Evaluation¶

This is a playground to test & debug the Digital Worker as we are developing, or at any time.

1. Conveniently test your Digital Worker at any time by clicking on the Evaluation button (arrow in circle) on the right sidebar
   1. For example: If a PDF normally submitted via email, you can upload a PDF instead and type out the relevant message in the chat option
2. While waiting for the agent to respond you can select Plan Flow (branching icon) to see the internal tool calls and sub agent handoffs
3. You can also see all your previous playground chats during evaluation by selecting Show Tabs (multi-line icon) on the right sidebar
4. Click the pencil icon next to any event in the list to view additional details