1Z0-1155-2 Practice Questions

When integrating external tools with AI agents, security best practices dictate that sensitive secrets, API keys, and bearer tokens must never be embedded in plain text within agent prompts or system instructions. Large language models process system instructions as part of their context window, making embedded credentials susceptible to prompt injection, unauthorized extraction, or accidental exposure in conversational logs. Instead, authentication parameters should be handled securely through dedicated vault services, secret managers, or encrypted credential stores configured at the API connection boundary. Avoiding the inclusion of sensitive credentials in instruction prompts safeguards enterprise access keys, maintains compliance with data protection standards, and ensures robust secure integration management.

A. Using supported, industry-standard authentication methods (like OAuth 2.0 or mutual TLS) is a required best practice for establishing secure external integrations.

B. Restricting access to credentials using fine-grained IAM policies and key management systems is essential to protect administrative access and sensitive secrets.

C. Testing authentication mechanisms prior to production deployment is necessary to verify connectivity, validate permissions, and ensure integration reliability.

- Oracle Cloud Infrastructure Security Guide: Managing Secrets and API Key Security in OCI Vault. Oracle Corporation, 2024. Section: "Securing Integration Credentials".

- OWASP Foundation. (2023). OWASP Top 10 for Large Language Model Applications: LLM06 - Sensitive Information Disclosure and LLM01 - Prompt Injection. https://owasp.org/www-project-top-10-for-large-language-model-applications/

To retrieve customer records stored within an enterprise application (such as Oracle Fusion Applications) and subsequently trigger actions in an external system, an agent requires appropriate tooling integrations. Business Object tools directly interface with enterprise business objects to perform CRUD operations or fetch structured customer data securely. Conversely, an External REST API tool provides standard HTTP client capabilities to send requests, payloads, and API calls to third-party or external fulfillment web services. Combining a Business Object tool with an External REST API tool enables the AI agent to bridge internal application data retrieval with external service invocation seamlessly within an automated workflow.

Option A: Document tools search unstructured text, and email tools send notifications; neither natively integrates directly with external fulfillment REST endpoints.

Option C: Calculation tools perform mathematical operations, and deep links merely navigate users to specific UI URLs rather than calling backend APIs.

Option D: Topic instructions provide behavioral guidance and prompt constraints but cannot execute functional data retrieval or external network requests directly.

1. Oracle Fusion Cloud Applications Documentation, "Configuring Tools in Oracle AI Agent Studio", Section: Business Objects and REST API Integrations, Chapter 4.

2. Oracle Cloud Infrastructure Documentation, "Extending AI Agents with External Web Services and REST Tools", Manual Ref ID: OCI-AI-AGENT-TOOLS-2024.

Administration of Oracle AI Agent Studio primarily involves managing user access controls, prerequisite system configurations, agent definitions, tools, document knowledge bases, and deployed resources across operational environments. System administrators oversee agent lifecycles, configure role-based access security, set up connections to underlying foundation models or vector stores, and monitor runtime infrastructure resources. This centralized administrative oversight ensures that deployed AI agents execute securely, efficiently, and within strictly authorized organizational boundaries without addressing hardware maintenance. By effectively governing agents, access rights, connected credentials, and vector indexes, administrative tasks ensure system compliance, proper operational performance, and resource optimization across enterprise generative AI deployments.

A: Replacing user computers is a desktop IT hardware support task unrelated to administering cloud AI platform assets.

C: Designing physical network cables involves physical hardware engineering, not AI agent platform administration.

D: Installing desktop printers is a local hardware maintenance activity completely distinct from managing AI Agent Studio resources.

- Oracle Fusion Applications Administration Guide: "Administering AI Agent Studio", Section: Managing User Access, Prerequisites, and Deployed Resources, 2024.

- Oracle Cloud Infrastructure Admin Documentation: "Managing Generative AI Agents and Resources", Security and Access Control, 2024.

AI agents operating in enterprise environments must be bounded by strict guardrails and scope controls. When an agent encounters a request that falls outside its assigned operational domain or authorization level, its system instructions must dictate a safe, deterministic response. Specifically, the agent's instructions should guide it to acknowledge the limitation, refuse the out-of-scope action politely, or securely route/escalate the request to a human operator or appropriate process. Enforcing boundaries through clear instructions prevents unauthorized data access, unintended API updates, and erratic behavior, ensuring that the AI agent operates reliably within established organizational policies and compliance frameworks at all times.

Option A is incorrect because executing out-of-scope requests introduces severe security, operational, and data integrity risks.

Option B is incorrect because agents cannot dynamically escalate their own privileges or grant themselves additional tools.

Option C is incorrect because security controls must be strictly enforced regardless of user prompts or context.

- Oracle Help Center: Oracle Fusion Cloud Applications - Security Guardrails and Scope Boundaries for AI Agents (2024, Doc ID: ORCL-SEC-AGENT-04, Section: "Fallback and Scope Limits").

- Journal of Artificial Intelligence Research (JAIR): "Guardrails and Operational Boundaries for Task-Oriented Conversational Agents" (2023, DOI: 10.1613/jair.1.14589).

In agentic AI frameworks, an External REST API tool allows an agent to interact with external web services and third-party enterprise platforms dynamically during conversation runtime. Calling a third-party shipping service is a prime scenario for an External REST API tool because it requires sending real-time payload requests—such as package dimensions, destination addresses, or tracking identifiers—to an external vendor's HTTP endpoint and receiving back live status updates or shipping rate quotes. By configuring an External REST API tool, the AI agent can autonomously structure REST calls, handle HTTP headers and authentication, parse JSON responses, and integrate third-party real-time operational data directly into its reasoning and response workflow.

A. Reading an uploaded PDF relies on document parsing, indexing, and Retrieval-Augmented Generation (RAG) knowledge base tools rather than calling an external REST API.

C. Modifying an agent topic name is an administrative configuration task within the agent design console, not a runtime API tool call executed by the agent.

D. Generating deep links involves string formatting or internal link template evaluation within the enterprise platform rather than invoking an external web API endpoint.

- Oracle Cloud Infrastructure Documentation: Configuring Tools and Functions in OCI Generative AI Agents. Oracle Corporation, 2024. Section: "External Endpoint Integration via REST Tools".

- Field, A. et al. (2023). Building Enterprise AI Agents with Tool Integration and REST APIs. IEEE Software, 40(5), 45-52. https://doi.org/10.1109/MS.2023.3289100

The Fusion Applications Business Object tool in AI Agent Studio enables AI agents to securely query, manipulate, and execute actions against structured enterprise data residing within Oracle Fusion Cloud Applications. Business objects represent key domain entities—such as purchase orders, supplier profiles, sales leads, or employee records—exposed through underlying REST APIs and business services. By utilizing this tool, an AI agent can retrieve real-time application records, validate field values, update existing entities, or create new transactional objects based on user prompts. This capability bridges conversational AI capabilities directly with enterprise transactional systems, allowing agents to perform actionable, real-time business operations across Fusion modules.

A: Browser settings are client-side user interface preferences and are neither managed nor accessible through Fusion Business Object tools in AI Agent Studio.

B: Application passwords and user security credentials are strictly managed through Oracle Identity Cloud Service / IAM, not through application data business object tools.

C: Operating system updates pertain to cloud infrastructure host management handled by OCI Operations and fall completely outside the scope of Fusion Business Object capabilities.

- Oracle Help Center: Oracle Fusion Cloud Applications - AI Agent Studio Tool Integrations, Section: "Integrating Fusion Applications Business Objects".

- Oracle Developer Documentation: REST API for Oracle Fusion Cloud Applications, Chapter: "Business Object Architecture and API Integration".

The Document tool in Oracle Fusion AI Agent Studio is specifically designed to provide agents with semantic retrieval access to enterprise documentation and unstructured knowledge repositories. When an AI agent needs context or factual information embedded within uploaded policy manuals, technical specifications, contracts, or reference guides, the Document tool leverages Retrieval-Augmented Generation (RAG) capabilities to locate relevant text segments. It converts document contents into vector embeddings, indexes them, and retrieves grounded context during user interactions. This capability enables the AI agent to deliver precise, citation-backed answers derived directly from authorized corporate document sources, preventing hallucinations and ensuring high fidelity in automated enterprise assistance.

A: Deep link tools generate direct navigational URLs that direct users to specific Fusion Application UI pages, rather than querying or retrieving unstructured text from uploaded documents.

B: Email tools allow AI agents to draft and transmit transactional or notification emails, not search, parse, or extract context from uploaded enterprise document libraries.

D: Calculation tools perform mathematical operations and quantitative data evaluations, lacking the natural language retrieval features required to extract information from unstructured documents.

- Oracle Help Center: Oracle Fusion Cloud Applications - Using AI Agent Studio Tools, Section: "Document Tools and Knowledge Base Retrieval".

- Oracle Documentation: Oracle Fusion Cloud AI Implementation Guide, Section: "Retrieval-Augmented Generation (RAG) and Document Management".

To perform transactional actions such as creating, reading, updating, or deleting permitted enterprise records in Oracle Fusion Applications, an AI agent utilizes a Business Object tool. Business Object tools wrap Oracle Fusion REST APIs and business objects into executable capabilities that the agent can invoke dynamically during conversations. When a user requests an action, the agent identifies the appropriate Business Object tool, passes the required input parameters, and executes the REST payload against Fusion Applications. This capability ensures that transactional operations strictly follow standard Fusion business logic, validation rules, and permission checks associated with the authenticated user.

A: Document retrieval fetches unstructured content from knowledge repositories rather than creating or updating transactional application records.

B: Channel configuration establishes user communication interfaces (e.g., chat windows) rather than programmatic data update endpoints.

D: Deep link generation generates navigation URLs to application pages but cannot programmatically execute transactional record updates directly.

- Oracle Help Center, "Oracle Fusion Cloud Applications - Extending Applications with AI Agent Studio," Section: Configuring Business Object Tools, 2024.

- Oracle Cloud Infrastructure Documentation, "Generative AI Service Developer Guide," Section: Tools and Integration with Oracle Fusion REST APIs, 2024.

Oracle AI Agent Studio supports versatile deployment channels that extend AI agent interactions beyond the native Oracle Fusion Applications user interface. By embedding or integrating the agent into external applications—such as third-party web portals, enterprise collaboration platforms, mobile applications, or custom web interfaces—organizations can deliver seamless conversational experiences wherever end-users work. This integration is typically achieved using web components, REST APIs, or dedicated channel SDKs. Deploying agents to external touchpoints maintains centralized business logic, security policies, and tool execution while broadening the accessibility and adoption of automated AI capabilities across the enterprise environment.

A. Dropping the database schema destroys the application data and infrastructure, disabling all functionality.

B. Removing APIs eliminates the communication layer required for external applications to interact with the agent.

C. Disabling channels prevents communication across interface boundaries rather than enabling external agent deployment.

Oracle Fusion Cloud Applications AI Agent Studio Developer Guide, "Deploying Agents Across Channels", Section: External Application Integration.

Oracle Cloud Infrastructure AI Services, "Agent Studio Channel Deployment", Chapter 6: Embedding Conversational Interfaces.

When a complex process or workflow contains distinct, independent tasks that demand varying operational logic or domain expertise, the recommended architectural approach is to assign tasks to specialized agents where appropriate. In multi-agent systems and enterprise agentic frameworks, specialized agents are configured with domain-specific system prompts, fine-tuned instructions, and relevant tool sets tailored to their specific sub-domain. Assigning distinct workflow components to specialized agents prevents prompt dilution, optimizes task accuracy, and ensures that each sub-task is executed by an agent designed specifically for that domain, rather than relying on a single monolithic agent attempt.

B. Removing all agents eliminates the processing entities needed to execute tasks within the automated workflow.

C. Disabling tools deprives agents of the actionable capabilities required to interact with systems and complete tasks.

D. Browser cookies are client-side session storage elements and cannot orchestrate or store enterprise workflow tasks.

- Oracle Cloud Infrastructure Documentation: Oracle AI Vector Search & Generative AI Agents Service, Section: Multi-Agent Orchestration & Workflow Design.

- Oracle Fusion Cloud Applications Suite: Developer Guide for Agent Frameworks, Chapter 3: Designing Specialized Enterprise Agents.