Oracle AI for Fusion Cloud CX: AI Agent Studio Developer Training Ed 1 — Lesson 2: Agent creation, instructions, actions, data access, prompts, orchestration, and Fusion CX integration

Learning Objectives

• Identify the configuration and data choices that shape Agent creation, instructions, actions, data access, prompts, orchestration, and Fusion CX integration.
• Run normal and boundary tests and record the outcome in working agent prototype and tool-call trace.
• Separate root cause from symptom when investigating an instruction conflict causing the agent to choose the wrong action.

Detailed Topics

• Agent creation
• Instructions
• Actions
• Data access
• Prompts
• Orchestration
• And Fusion CX integration
• Decision ownership, prerequisites, and acceptance evidence for Agent creation, instructions, actions, data access, prompts, orchestration, and Fusion CX integration
• Relevant access, data, integration, performance, recovery, and support considerations

Instructor Demonstration

The instructor configures and tests Agent creation, instructions, actions, data access, prompts, orchestration, and Fusion CX integration within a scenario aimed at a safe and useful Fusion Cloud CX AI agent designed, grounded, tested, deployed, monitored, and governed for business service. The walkthrough contrasts a valid setup with a flawed one and uses working agent prototype and tool-call trace to make the difference explicit.

Hands-On Exercise

Configure and test Agent creation, instructions, actions, data access, prompts, orchestration, and Fusion CX integration with representative records. Use “an instruction conflict causing the agent to choose the wrong action” as the troubleshooting challenge, compare competing hypotheses, retest only the affected path, and complete working agent prototype and tool-call trace for peer review.

1. Confirm scope, access, sample records, and acceptance criteria.
2. Perform the configuration or transaction using a controlled sequence.
3. Inspect outputs and downstream effects rather than relying on screen messages alone.
4. Close with a support note, preventive action, and reproducible validation step.

Scenario Analysis

• Clarify the business trigger, affected roles, required records, and acceptable result for Agent creation, instructions, actions, data access, prompts, orchestration, and Fusion CX integration.
• Compare at least two implementation or correction choices and identify which evidence would distinguish them.
• Decide how working agent prototype and tool-call trace should be reviewed, approved, retained, and handed to the next owner.

Troubleshooting Guidance

When results from Agent creation, instructions, actions, data access, prompts, orchestration, and Fusion CX integration are unexpected, first reproduce the issue with the same role, record, date, and transaction context. Inspect working agent prototype and tool-call trace, then test whether “an instruction conflict causing the agent to choose the wrong action” explains the evidence before changing configuration. Retest the corrected path and one unaffected path to guard against regression.

Validation Checklist

• The scope of Agent creation, instructions, actions, data access, prompts, orchestration, and Fusion CX integration is clear, including what is deliberately excluded from the test.
• Evidence in working agent prototype and tool-call trace supports each material decision and outcome.
• A negative test demonstrates that controls reject or flag an invalid condition.
• Open risks, owners, due actions, and production safeguards are documented.

Lesson Outcome

Learners can apply Agent creation, instructions, actions, data access, prompts, orchestration, and Fusion CX integration, explain the decisions made, and defend the result with working agent prototype and tool-call trace.