SAP Agent¶

SAP Agent is a third-party agent hosted on the SAP platform. These enterprise-grade AI agents leverage SAP’s Business AI to automate workflows, analyze real-time business data, assist in financial operations, and deliver contextual responses powered by organizational knowledge and SAP systems integration.

The AI Refinery SDK enables effortless integration with a user's tailored SAP Agent through the SAPAgent class, unlocking the full potential of the SAP Agent to boost performance and expand functionalities within the AI Refinery platform.

Creating SAP Agent ¶

Users can develop a SAP Agent using a tailored approach on the SAP platform. If you are interested in creating an agent, please reach out to us and we will provide guidance based on the customized solution. Once the agent is created, you can retrieve the endpoint URL, which refers to the location where the agent operates.

Onboarding SAP Agent¶

To use the SAP agents through our AI Refinery SDK, users need the following parameters:

Variable	Description	Required
`url`	The SAP host URL, which is an identifier of where your SAP agents live.	Yes
`contexts`	Allows for the provision of additional information during communication with the `SAP Agent`.	No

Workflow Overview¶

The workflow of the SAPAgent class consists of four components:

Initialization: The user must create the SAP agent on the SAP AI agent platform, and register it in the AI Refinery using the corresponding SAP host URL.
Sending a Query: A user query is forwarded from AI Refinery to the SAP Agent operating on the SAP platform.
SAP-Side Processing: The SAP Agent decides whether to utilize predefined tools, knowledge bases, or other resources and handles the query within the SAP platform.
Receiving and Parsing the Response: The SAP Agent returns the processed results to AI Refinery, which then parses them to derive the final answer.

Usage and Quickstart¶

To quickly set up an AI Refinery project with a SAPAgent, the user should first create their own SAP Agent in the SAP platform as explained above. Once the agent is ready, use the YAML configuration template below to integrate it into the AI Refinery project. Specifically, ensure the following configurations are included:

Add a utility agent with agent_class: SAPAgent under utility_agents.
Ensure the agent_name you chose for your SAPAgent is listed in the agent_list under orchestrator.

Template YAML Configuration of `SAPAgent`¶

See the YAML template below for the SAPAgent configuration.

orchestrator:
  agent_list:
    - agent_name: "SAP Orders Agent"

utility_agents:
  - agent_class: SAPAgent
    agent_name: "SAP Orders Agent"
    agent_description: "The SAP Orders Agent can answer questions regarding orders recorded in the SAP database."
    config:
      url: <SAP Host URL> # Required: URL specifying where the SAP host is located.
      contexts: # Optional: additional agent contexts
        - "date"
        - "chat_history"