CCNA Agentic Solutions Questions

Option D is correct because the weather API needs to trust the managed identity's token, which requires configuring the API as a federated identity credential or using an app registration. Option A is wrong because the resourceId is correctly formatted. Option B is wrong because the model provider is valid.

Option C is wrong because the endpoint is not necessarily incorrect.

Option C is correct because the DialogSet manages the dialog stack and uses state to persist between turns. Option A is wrong because IBot is the main interface but does not manage dialogs directly. Option B is wrong because StateMiddleware is part of the Bot Framework but not dialog-specific.

Option D is wrong because ActivityHandler handles activities but not dialog management.

Threads in Azure AI Agent Service are designed to store conversation history and maintain context. Memory is a generic term, not specific to the service. Vector stores are for embeddings.

Sessions are for containerized apps.

Using Azure Key Vault references in the function code is the secure way to retrieve secrets at runtime without exposing them. Hardcoding is insecure. Environment variables may be exposed.

Passing in the prompt leaks credentials.

Option D is correct because 'powerAutomateFlow' is the action type for executing Power Automate flows. Option A is wrong because 'openApi' is for REST APIs. Option B is wrong because 'httpRequest' is for direct HTTP calls.

Option C is wrong because 'function' is for custom code functions.

Option A is correct because the function key is retrieved from the default key, but the function might require a specific key. Option B is wrong because the function key is correctly passed as a header. Option C is wrong because the endpoint is valid.

Option D is wrong because the model name is valid.

Option C is correct because 'Transfer conversation' automatically passes the full transcript to the human agent. Option A is incorrect because 'Create a ticket' does not transfer the conversation. Option B is incorrect because 'End conversation' terminates without escalation.

Option D is incorrect because 'Start a new topic' does not include history.

OAuth 2.0 authorization code flow with PKCE is the recommended approach for delegated permissions in native apps that need to act on behalf of a user. Client credentials flow is for app-only access. Implicit flow is deprecated.

Device code flow is for devices without a browser.

Code Interpreter can execute SQL queries via Python. Function calling can be used to invoke custom database operations. Knowledge base is for search, not write.

KQL is for Azure Data Explorer.

Option D is correct because it addresses all requirements: on-premises data gateway for NTLM-secured API, custom connector for OAuth 2.0, generative answers with SharePoint for multi-language policies, and SSO with Microsoft Entra ID. Option A is wrong because Power Automate flows cannot directly use NTLM without a gateway. Option B is wrong because Azure Functions cannot directly access on-premises without a gateway.

Option C is wrong because Logic Apps need a gateway and custom connectors are better for OAuth 2.0.

Option B is correct because summarizing the data reduces its size while preserving key information. Option A is incorrect because truncation may lose critical data. Option C is incorrect because splitting into multiple calls increases latency and tokens.

Option D is incorrect because marking as sensitive does not solve the size issue.

Option C is correct because generative answers use large language models to understand natural language and handle complex queries. Option A is wrong because topics are for structured conversations. Option B is wrong because entities extract specific data but not understand intent.

Option D is wrong because Power Automate flows are for backend actions, not understanding.

Option C is correct because client certificate authentication is used for certificate-based auth. Option A is wrong because OAuth 2.0 is token-based. Option B is wrong because API Key is key-based.

Option D is wrong because Windows Authentication is for on-premises.

The Code Interpreter tool allows the agent to execute code, including SQL queries, against data sources. Function calling is for invoking custom APIs, not direct SQL execution. Grounding with Bing is for web search.

KQL is for Azure Data Explorer.

Option C is correct because a sequential flow with multiple question nodes ensures the user provides information step-by-step. Option A is wrong because a single question node will not handle multiple pieces of information. Option B is wrong because adaptive cards are for rich interactions, not step-by-step data collection.

Option D is wrong because generative answers are for free-form text, not structured collection.

Option A is correct because the Teams channel in Copilot Studio allows authentication configuration to restrict access to specific tenants. Option B is wrong because turning off authentication would allow anyone. Option C is wrong because the Direct Line channel is for custom apps, not Teams.

Option D is wrong because the Web channel would not enforce Teams-specific authentication.

Option D is correct because grounding with retrieved documents reduces hallucination. Option B is correct because limiting source documents reduces noise. Option A is wrong because increasing temperature increases randomness.

Option C is wrong because system message alone doesn't guarantee fidelity. Option E is wrong because chunk size optimization is about performance, not contradiction.

Option C is correct because Azure API Management with on-premises data gateway can securely expose on-premises APIs to the cloud and handle Windows Authentication. Option A is wrong because Azure Functions would need a gateway to connect to on-premises. Option B is wrong because Logic Apps also need a gateway.

Option D is wrong because VPN is not a service for API management.

Options A and D are correct. Authentication ensures the user's identity is verified. Topic-level security restricts access to specific topics based on user roles.

Option B is wrong because audit logs are for monitoring, not access control. Option C is wrong because bot-level authentication does not restrict topics. Option E is wrong because data loss prevention does not control access within the agent.

Option B is correct because when the agent triggers the flow, it uses the agent's identity, which may not have permissions to the CRM. Option A is wrong because connector permissions are typically configured once. Option C is wrong because the flow runs at runtime, not after sign-in.

Option D is wrong because the flow works manually, so the connection is valid.

Option D is correct because language analyzers in Azure AI Search handle language-specific text processing. Option A is wrong because semantic search improves relevance, not language handling. Option B is wrong because scoring profiles rank results.

Option C is wrong because synonyms expand queries, not handle languages.

Option D is correct because 'Trigger phrases' in Copilot Studio are used to match user utterances to topics. Option A is incorrect because 'Entities' extract specific information, not intent. Option B is incorrect because 'Variables' store data.

Option C is incorrect because 'Actions' perform operations.

Options A and D are correct. A: Entities allow extraction of specific data like order numbers. D: Variables store the collected information for later use.

Option B is incorrect because trigger phrases map intent, not capture data. Option C is incorrect because Power Automate flows process data but are not for initial capture. Option E is incorrect because topics handle conversation flows, not data capture.

Option B is correct because the managed identity needs Graph API permissions assigned. Option A is wrong because the endpoint is valid. Option C is wrong because the resourceId is correctly specified.

Option D is wrong because the apiVersion is a preview version but that doesn't cause authorization errors.

Option A is correct because OAuth 2.0 with delegated permissions allows the agent to act on behalf of the user after consent. Option B is wrong because application permissions are for app-only access, not user delegation. Option C is wrong because API key does not support user context.

Option D is wrong because certificate-based auth is also app-only.

Option D is correct because UserState with Cosmos DB storage provides a scalable solution for storing user preferences across conversations. Option A is wrong because ConversationState is scoped to a single conversation. Option B is wrong because Blob Storage is for large objects, not key-value state.

Option C is wrong because Bot State Service is deprecated.

Agent Handoff is a built-in feature in Microsoft Foundry that allows one agent to pass control to another agent along with context. Custom API integration is possible but not the designed approach. Shared memory is for state management, not handoff.

Sequential orchestration is manual.

Option C is correct because the agent's response might be based on the topic's condition, not the flow output. Option A is wrong because if the flow fails, the agent wouldn't get a response. Option B is wrong because parsing JSON is standard.

Option D is wrong because the agent doesn't require a specific variable type.

Option B is correct because increasing the number of results gives the agent more context to choose from, reducing irrelevant returns. Option A is incorrect because decreasing the search score threshold may include more irrelevant results. Option C is incorrect because semantic ranking improves relevance but may not help if the initial retrieval is poor.

Option D is incorrect because changing the analyzer affects tokenization, not relevance directly.

Option C is correct because the security section uses api_key authentication, which is not suitable for calling Microsoft Entra ID-protected functions. Option A is incorrect because the model is supported. Option B is incorrect because the functions are properly defined.

Option D is incorrect because conversation_starters do not affect authentication.

Option D is correct because 'code_interpreter' action type allows executing Python code in a sandbox with package installation. Option A is wrong because 'function' is for calling Azure Functions. Option B is wrong because 'httpRequest' is for HTTP calls.

Option C is wrong because 'openApi' is for REST APIs.

Option D is correct because a token bucket with a shared counter across all agent instances ensures global rate limiting without dropping requests unexpectedly. Option A is incorrect because 'retry-after' headers require the agent to handle 429 responses, which can degrade UX. Option B is incorrect because exponential backoff can cause delays.

Option C is incorrect because a fixed delay per request does not adapt to varying traffic.

Option A is correct because a Power Automate flow can be triggered from a topic and send an email using the user's email address. Option B is wrong because the email channel is for receiving emails. Option C is wrong because Copilot Studio does not have a native email action.

Option D is wrong because adaptive cards are for in-chat interactions, not sending emails.

Option B is correct because if the flow runs in test but not in published mode, the most common issue is that the flow connection is not shared with the agent. Option A is wrong because permissions are likely correct if test works. Option C is wrong because the flow is already created.

Option D is wrong because the trigger is correct if test works.

Parallel tool execution reduces latency by running independent tool calls concurrently. Adding more tools increases load. Increasing token limit does not help with tool call latency.

Reducing thread history might affect context but not tool call speed.

Option A is correct because OAuth 2.0 with on-behalf-of flow allows the agent to authenticate to Microsoft Entra ID and then access on-premises resources via an API. Option B is wrong because API key is not suitable for user-specific actions. Option C is wrong because certificate-based authentication is not typical for this scenario.

Option D is wrong because basic authentication is insecure and deprecated.

Options B and D are correct. B: The agent must have a managed identity assigned to authenticate to resources like Microsoft Graph and Azure SQL Database. D: The agent's code must use the managed identity to acquire tokens, typically via DefaultAzureCredential.

Option A is incorrect because connection strings expose secrets. Option C is incorrect because RBAC permissions are assigned to the managed identity, not the agent's service principal directly. Option E is incorrect because connection strings are not used with managed identities.

Option B is correct because the skill might not be passing the API key as a header. Option A is wrong because the function is reachable. Option C is wrong because the key might be correct but not passed.

Option D is wrong because the function URL might be correct.

Option A is correct because without tools, the agent can only respond based on its instructions and model, limiting its capabilities. Option B is incorrect because the agent can still start conversations. Option C is incorrect because the agent can still be deployed.

Option D is incorrect because the agent can still be invoked.

Option C is correct because 'Authentication' in Copilot Studio allows you to require Microsoft Entra ID sign-in with MFA. Option A is incorrect because 'Power Automate flow' can call MFA but is not the primary configuration. Option B is incorrect because 'Variable' does not enforce authentication.

Option D is incorrect because 'Topic trigger' initiates the topic but does not enforce authentication.

Option B is correct because topic-level security with authentication is required to restrict access to the handbook content and ensure the agent only responds to relevant queries. Option A is wrong because generative answers alone may pick up unrelated content. Option C is wrong because bot-level security does not restrict topics.

Option D is wrong because disabling fallback does not prevent the agent from answering unrelated queries using generative answers.

Data residency requires storing and processing data within the EU region. Using Azure regions in the EU, configuring data storage in those regions, and enabling data encryption at rest are key compliance measures. Global regions may violate residency.

Disabling logging may hinder auditing.

Option D is correct because the ParseJSON function in Copilot Studio converts a JSON string into a Power Fx record that can be used to access properties. Option A is wrong because the agent does not have a generic JSON parser built-in. Option B is wrong because Power Automate can parse JSON, but the agent needs to parse it in Copilot Studio.

Option C is wrong because the agent cannot use C# directly.

Option C is correct because middleware runs on every incoming message and can translate before the bot logic runs, minimizing latency by avoiding multiple round trips. Option A is wrong because translating in the activity handler would still be in the pipeline but more coupled. Option B is wrong because the language detection API is separate.

Option D is wrong because client-side translation is not possible in this scenario.

Option D is correct because a SequentialPlanner combined with a stepwise strategy ensures the agent only invokes the API when the context explicitly requires it, reducing token waste. Option A is incorrect because AutoInvoke with greedy planning may over-call APIs. Option B is incorrect because ManualInvoke requires user intervention, defeating autonomy.

Option C is incorrect because ParallelPlanner may invoke multiple APIs concurrently, increasing token usage.

Option B is correct because a low maximum context length truncates earlier conversation turns, causing the agent to lose context and repeat responses. Option A is incorrect because the token limit per request affects the length of individual responses, not repetition. Option C is incorrect because the Azure OpenAI API does not recycle context unless explicitly configured.

Option D is incorrect because changing temperature affects randomness, not repetition due to context loss.

Options B, C, and E are correct. B: Providing few-shot examples helps the planner understand complex tasks. C: Creating higher-level composite functions reduces the number of steps the planner needs to orchestrate.

E: Allowing the planner to ask clarifying questions improves accuracy. Option A is incorrect because a simple prompt reduces capability. Option D is incorrect because limiting functions restricts the planner.

Option A is correct because a supervisor agent with a shared context store allows coordination and handoff. Option B is incorrect because separate independent agents lack coordination. Option C is incorrect because sequential chaining limits flexibility.

Option D is incorrect because a single monolithic agent defeats the purpose of specialization.

Option B is correct because 'file_search' is disabled in the template, so the agent cannot search uploaded documents. Option A is incorrect because code_interpreter is enabled. Option C is incorrect because the function is correctly defined.

Option D is incorrect because the model is valid.

Knowledge base tool (or file search / grounding) is used to connect to Azure AI Search index. Code Interpreter is for code execution. Function calling is for APIs.

KQL is for Azure Data Explorer.

Option C is correct because Microsoft Graph API requires OAuth 2.0 tokens. Option A is wrong because API Key is not supported. Option B is wrong because Basic auth is deprecated.

Option D is wrong because client certificates are for server-to-server but OAuth is standard.

Option B is correct because Azure Monitor collects telemetry from Azure resources. Option A is wrong because Azure Log Analytics is part of Azure Monitor, not a separate service for monitoring. Option C is wrong because Application Insights is for application performance, not infrastructure monitoring.

Option D is wrong because Microsoft Sentinel is a SIEM, not general monitoring.

Option D is correct because document-level security filters in Azure AI Search enforce permissions. Option A is wrong because disabling public access is not granular. Option B is wrong because managed identity alone doesn't filter documents.

Option C is wrong because MFA is authentication, not authorization.

Option A is correct because HTTP with OAuth2 client credentials is the appropriate way to authenticate with an external API that uses OAuth 2.0. Option B is wrong because API Key is not OAuth. Option C is wrong because Power Automate with native CRM connector might not support OAuth client credentials.

Option D is wrong because AI Builder is not for API authentication.