AI Agent: Quote to Purchase Requisition Chat Assistant

In update 25D, we introduced the AI Agent: Quote to Purchase Requisition Assistant feature, which parses emails containing supplier quotations and automatically converts them into purchase requisitions. Now, this automatic conversion can also be initiated directly from chat, allowing users to upload supplier quotations within the chat interface.

Quote to Purchase Requisition Chat Assistant

Quote to Purchase Requisition Chat Assistant

Steps to enable and configure

To enable the Quote to Purchase Requisition Chat Assistant AI Agent, follow these steps:

  1. Create and publish an instance of the seeded "Quote to Purchase Requisition Chat Assistant".
    Navigate to AI Agent Studio, create an instance of the seeded “Quote to Purchase Requisition Chat Assistant”. Publish your instance.
    There are two ways to create your instance: 
    • Copy Template: Automatically copies all underlying artifacts from the seeded agent team and adds a suffix you specify. You can edit the agent team’s settings, agents, tools, and topics.
    • Use Template: Takes you through a step-by-step configuration of each artifact in the agent team.
  2. Create a new Guided Journey and add an agent task of the type Workflow Agent for the new agent you just created in AI Agent Studio. 
  3. Using Oracle Visual Builder Studio, add the Guided Journey to the Redwood page from where you want your users to have access to it. 
  4. Run the Prepare the Purchasing Category Suggestion Model (POR_PREPARE_PURCHASING_CATEGORY_SUGGESTIONS_MODEL) scheduled job to create the classifier that will classify (suggest a category) lines based on their item description.
    • The training will be based on agreements and purchase orders.
    • Schedule the scheduled job to run at regular intervals (this way the training or classifier is updated as more data becomes available).
  5. Run the Self-Service Procurement Vector Index Generator AI agent with the EntityName as PURCHASING_UOM. This loads Fusion UOMs into the Vector DB which is then used to derive UOMs from the supplier quotations
    • Navigate to AI Agent Studio > Agent teams and select the Published tab.
    • Search for Self-Service Procurement Vector Index Generator and click the View icon.
    • Click the Run icon.
    • Select Rest for the trigger type.
    • Enter PURCHASING_UOM for EntityName
    • Select Start Test.
      Self Service Procurement Vector Index Generator

      Self-Service Procurement Vector Index Generator

       


      Input for Procurement Artifact Vector Generator

      Input for Self-Service Procurement Vector Index Generator

Tips and considerations

  • The supplier quotation file size must not exceed 50 MB.
  • Only text-based PDF attachments are supported, image-based (scan to PDF) files aren't supported.
  • For your end-users to see the agent in Guided Journey,
    • you must specify a role in the security tab of your Agent Team instance 
    • end-users must be granted this role

Key resources

Access requirements

To access the Oracle AI Agent Studio for Fusion Applications and manage PRC AI agents, users must be assigned a configured job role that contains these duty roles:

  • PRC Intelligent Agent Management Duty (ORA_PO_PRC_AI_AGENT_MANAGEMENT_DUTY)
  • PRC Intelligent Agent Management Duty (ORA_PO_PRC_AI_AGENT_MANAGEMENT_DUTY_HCM)
  • Fai Genai Agent PRC Administrator Duty (ORA_DR_FAI_GENERATIVE_AI_AGENT_PRC_ADMINISTRATOR_DUTY)

To interact with AI agents in product pages, users must be assigned a configured job role that contains this duty role:

  • Fai Genai Agent Runtime Duty (ORA_DR_FAI_GENERATIVE_AI_AGENT_RUNTIME_DUTY)

The Chat Assistant is available via Guided Journey. To access Guided Journey, users must be assigned a configured job rule that contains

  • Access HCM Common Components (HRC_ACCESS_HCM_COMMON_COMPONENTS_PRIV)
  • Use REST Service – Guided Journeys Read Only (ORA_PER_REST_SERVICE_ACCESS_GUIDED_JOURNEYS_RO) 

Also, refer to the tip in the Tips & Considerations section.

In the Security Console, filter by Roles and Permission Groups to find this duty role.

To allow users to interact with agents, you must also enable permission groups in the Security Console on those users’ configured job roles that contain the Fai Genai Agent Runtime Duty role. You can enable permission groups when you manage the basic information of your configured job roles.

Users’ configured job roles must also contain privileges that allow access to the pages where AI agents are enabled.

Users who are assigned a configured job role that contains this existing duty role can schedule Agent Teams:

  • Fai Batch Job Manager Duty (ORA_DR_FAI_BATCH_JOB_MANAGER_DUTY).

See Access Requirements for AI Agent Studio (for administrators) and How can I give users access to AI agents (for end-users) for more information.