Microsoft Certified: Multi-Agent AI Solutions Expert practice questions
Microsoft · AI-500 · 300 questions
Original practice questions for Microsoft Certified: Multi-Agent AI Solutions Expert.
This course contains the use of artificial intelligence.
Practice Quizzes
Test your knowledge with standard 20-question practice sets.
Quiz 1
Quiz 2
Quiz 3
Quiz 4
Quiz 5
Quiz 6
Quiz 7
Quiz 8
Quiz 9
Quiz 10
Quiz 11
Quiz 12
Quiz 13
Quiz 14
Quiz 15
Browse by Domain
Study specific topics at your own pace.
Architect multi-agent solutions (15–20%) · 60 questions
- Contoso Insurance is designing a multi-agent system in Microsoft Foundry to process complex commercial claims. Business goals include document intake, fraud risk scoring, policy coverage matching, reserve estimation, and adjuster handoff. The solution architect must decompose these goals into agents and workflows. Which design approach best matches production multi-agent architecture practices?
- A healthcare network designs a multi-agent care-coordination solution. Agents can draft discharge summaries, propose medication changes, and schedule follow-ups. Clinical leadership requires that any medication change recommendation be reviewed by a licensed clinician before it is written to the electronic health record (EHR). Which architectural control best satisfies this requirement?
- Fabrikam Logistics is specifying agent personas for a fleet-operations multi-agent system. One agent negotiates carrier rates via external APIs; another summarizes SLA breaches for executives. Security requires least privilege and clear behavioral boundaries. Which specification set is most appropriate for the rate-negotiation agent?
- An enterprise customer-support multi-agent solution must remember the current ticket thread, share selected facts across specialist agents during a session, and retain resolved-case embeddings for future similar incidents without mixing tenants. Which memory architecture best addresses these needs?
- Adventure Works is integrating a research agent that must call internal knowledge tools and also collaborate with a partner-built planning agent hosted outside Adventure Works’ subscription. Architects must choose communication approaches. Which protocol selection is most appropriate?
- A bank is designing a Zero Trust multi-agent investment-research platform on Azure. Compliance demands that a compromised research agent must not be able to invoke trade-execution tools used by a separate trading agent. Which architectural specification best implements this requirement?
- Wide World Importers needs a multi-agent order-exception handler that spikes during holiday peaks, must recover from agent failures, and must control inference cost. Architects are selecting compute for orchestration workers and tool hosts. Which approach best balances scalability, reliability, operational efficiency, security, and cost?
- Northwind’s multi-agent underwriting system fails intermittently when three specialist agents collaborate. Engineers need to reconstruct which agent reasoned what, which tools ran, and how tokens were consumed across services. Which observability components should the architect specify?
- A product team is assigning models to agents in a multi-agent software delivery assistant. Tasks include: (1) high-stakes architecture critique requiring deep reasoning, (2) bulk classification of thousands of low-risk backlog items, and (3) fast intent routing in the orchestrator. Which model-matching strategy is most appropriate?
- Litware is designing a multi-agent workplace assistant that can draft emails, propose calendar changes, and call internal HR tools. Legal and UX require transparency, user control, and minimization of harmful or over-autonomous behavior. Which design package best supports human-AI experience (HAX) and Responsible AI principles?
- You are designing a multi-agent solution for corporate travel. A supervisor must route requests to specialized agents (flights, hotels, policy compliance) that may run concurrently, then merge results for the user. Which logical orchestration pattern best fits?
- Stakeholders want a refund agent that can draft refund recommendations but must never execute refunds above $500 without human approval. How should you specify the agent’s design?
- A multi-agent legal research system needs turn-by-turn conversational context for the active session and durable recall of matter-specific facts across weeks, with tenant isolation between law firm clients. Which memory architecture should you specify?
- You must integrate multiple internal APIs and Azure services as tools for several agents, with a standard protocol for tool discovery and invocation rather than one-off SDKs per agent. Which integration approach best fits modern multi-agent design on Azure?
- A multi-agent compliance workflow flags ambiguous policy cases. Business requires humans to approve, override, or escalate before external communications are sent. How should you design the workflow controls?
- A regulated multi-agent architecture review asks which components must be specified to prevent lateral movement and map controls to compliance frameworks. Which answer is most complete?
- During architecture design, the ops team insists the multi-agent system must support reproducible debugging of agent reasoning and cross-service traces after incidents. Which components should you specify up front?
- You are selecting models for a multi-agent system: a lightweight classifier agent routes intents; a deep reasoning agent handles complex multi-step plans; a vision agent reads PDF screenshots. What architectural principle should guide model assignment?
- A collaborative multi-agent design needs (1) per-user session state, (2) shared team workspace state for agents cooperating on one case, and (3) long-term semantic memory across cases, each with different lifecycle and isolation rules. Which persistence approach should you specify?
- Executives require that a customer-facing multi-agent assistant expose uncertainty, allow users to correct the agent mid-flow, and avoid deceptive anthropomorphism, while logging consent for data use. Which architectural focus addresses these requirements?
- A retailer wants an autonomous system that, for each new product launch, researches competitors, drafts marketing copy, checks brand compliance, and schedules social posts. Stakeholders describe this as one mega-agent that “does everything.” As the solution architect, what is the most appropriate first logical-architecture step?
- You are specifying agents for an internal IT multi-agent helpdesk. A password-reset agent may complete routine resets automatically, but license-purchase and firewall-change requests must pause for engineer approval. What should you define for each agent during logical design?
- An enterprise multi-agent platform must let planning agents discover and call external tools from multiple teams and must also allow a partner’s independently hosted agent to collaborate on shared cases. Which protocol selection best matches these integration needs?
- A multi-agent investigative workflow needs turn-by-turn dialogue context, a shared case file that several agents update during a session, and durable knowledge of prior cases for semantic recall across weeks. Tenant isolation and lifecycle policies are mandatory. Which memory architecture should you specify?
- You are selecting technology components for a multi-agent system that will process regulated financial data. Leadership mandates Zero Trust: no implicit trust between agents, minimized lateral movement, and mappable compliance controls. Which component specification is most aligned?
- During design review, operations requires the ability to reconstruct why a multi-agent loan decision path produced a given outcome, correlate spans across orchestrator and tools, and replay problematic runs for debugging. Which observability components should you specify?
- Your multi-agent architecture includes (1) a planner that breaks ambiguous goals into steps, (2) a high-volume classifier that routes tickets into ten categories, and (3) a code-generation specialist for infrastructure-as-code snippets. How should you match tasks to model capacities?
- You are designing the human-AI experience for a multi-agent legal research assistant used by junior associates. Responsible AI stakeholders require transparency about agent actions, ability to intervene, and clear limits on autonomy for filing-related recommendations. Which design elements should you include?
- A multi-agent batch analytics job can spawn dozens of short-lived worker agents overnight, then drop to near-zero load by day. Requirements emphasize elastic scale, reliability, operational efficiency, security isolation, and cost control. Which compute specification approach is most appropriate?
- A new multi-agent squad is forming across three time zones. Developers currently install conflicting Python packages locally, lack shared AI coding instructions, and onboarding takes weeks. You must specify developer tools and environments that support a healthy SDLC. What should you propose?
- Contoso Insurance is designing a multi-agent claims intake system in Microsoft Foundry. The business goal is: "Accept a first notice of loss (FNOL), classify claim type, gather supporting documents, and either auto-approve low-risk claims or route complex claims to a human adjuster." The architect must decompose this goal before selecting models or hosting. Which decomposition best matches production multi-agent design practice?
- Fabrikam Finance is designing three agents for invoice processing: IntakeAgent (parse invoices), MatchingAgent (3-way match PO/receipt/invoice), and PaymentAgent (initiate payment). Compliance requires that no agent may release payments above $25,000 without a finance manager approval. How should the architect specify autonomy levels and behavioral guidelines?
- Northwind Traders is integrating an existing inventory specialist agent built by another team with a new customer-service orchestrator. The inventory agent already exposes skills over a standard agent-to-agent interface. The orchestrator also needs to call a warehouse REST API for stock counts. Which protocol assignment is most appropriate?
- Adventure Works is designing memory for a multi-agent project planning suite: a Planner agent, a Risk agent, and a Scheduler agent collaborate on long-running programs spanning weeks. Users expect same-session continuity, shared team decisions during a sprint, and recall of prior project outcomes months later with strict tenant isolation. Which multi-tier state design should the architect specify?
- A regulated healthcare provider is deploying a multi-agent clinical documentation system on Azure. Security architects mandate Zero Trust for agentic systems: each agent must have its own identity, tool access must be least privilege, and a compromised summarization agent must not be able to move laterally into the e-prescribing tool path. Which design best meets these requirements?
- Wide World Importers is selecting models for a multi-agent supply-chain suite in Microsoft Foundry. Tasks include: (1) high-volume SKU classification from short structured fields, (2) multi-step negotiation email drafting with careful tone, and (3) overnight batch reconciliation of millions of line items with strict cost caps. Which model capacity matching approach is most appropriate?
- Tailwind Toys is launching a customer-facing multi-agent shopping assistant. Responsible AI and HAX (human-AI experience) requirements include: users must understand when AI is acting, high-impact returns/refunds need human confirmation, and the system must avoid dark patterns that pressure purchases. Which architectural controls best satisfy these requirements?
- Litware's platform team must specify observability for a multi-agent loan underwriting system spanning orchestrator, credit-check agent, and document agent across Azure services. Incidents require reconstructing why an agent chose a tool and replaying runs for debugging. Which observability component set should they specify?
- City Power & Light is choosing compute for a multi-agent outage-response system. Requirements: burst scale during storms, isolate long-running planner jobs from low-latency chat agents, maintain reliability with health probes, and control cost during calm weather. Which compute specification is most appropriate?
- A multi-agent engineering team at Contoso wants consistent local development for Microsoft Agent Framework projects, including shared Python dependencies, VS Code extensions, CLI tooling, and checked-in AI coding instructions for agent patterns. Which technology-component specification best supports the SDLC?
- An enterprise wants a multi-agent customer operations system where a central planner assigns work to billing, shipping, and returns specialists, enforces shared policies, and consolidates a single customer-facing answer. Specialists should not negotiate job assignment among themselves. Which logical orchestration pattern best matches this requirement?
- A bank designs a multi-agent lending assistant. Soft information collection and document checks may run autonomously, but any recommendation to approve credit above a regulatory threshold must include a human decision before systems of record update. How should the architect specify autonomy and control loops?
- An architect must design memory for a multi-agent consulting workspace. Agents need the last few turns for immediate coherence, a team-shared board of intermediate findings during a multi-hour engagement, and durable client preferences that survive across sessions for the same tenant. Which memory architecture best fits?
- A multi-agent media pipeline includes (1) language detection on short strings, (2) multi-hop legal reasoning over contracts, and (3) creative storyboard prose. The architect must assign model families without overspending. Which matching strategy is most appropriate?
- Before implementation, an architect must document how a triage agent, a medical-coding agent, and a billing agent should behave in a hospital revenue-cycle multi-agent system. What specification elements are most critical to define for each agent?
- Two business units already run production specialist agents on different stacks. A new enterprise orchestrator in Microsoft Foundry must invoke those existing agents as tools/services without rewriting them from scratch, and must also expose enterprise knowledge sources through a standard tool interface. Which protocol approach should the architect specify?
- A global retailer needs multi-agent compute that scales for Black Friday traffic spikes, isolates noisy neighbor batches, meets reliability targets, and controls cost during quiet hours. Which compute specification direction best balances these concerns?
- A government contractor must specify Zero Trust components up front for a multi-agent casework system handling controlled unclassified information. Which set of components belongs in the logical architecture?
- An architect is selecting observability components for a multi-agent fraud-investigation system that must support reproducible debugging when investigators challenge an automated conclusion weeks later. Which capability set should be specified?
- A public-sector multi-agent benefits assistant will converse with vulnerable citizens. The architect must design human-AI experience (HAX) and responsible AI controls into the workflow, not bolt them on later. Which design choice best reflects that mandate?
- A retailer wants a multi-agent solution that handles “resolve a late shipment”: detect delay, check inventory alternatives, propose options to the customer, and create a replacement order only after confirmation. How should you decompose the goal into agents and tools at the logical architecture level?
- You are selecting a communication topology for five specialist agents that must stay loosely coupled, with a single place to enforce routing policy, logging, and escalation. Peer agents should not freely call each other. Which logical architecture fits best?
- A legal firm wants a drafting agent that may propose contract clause language but must never send email to counterparties or accept settlement terms. Which specification elements belong in the agent’s persona and autonomy design?
- An enterprise multi-agent assistant needs (1) turn-by-turn dialogue context, (2) shared scratch state for the agent team during a case, and (3) long-term semantic memory of resolved cases—with tenant isolation. Which memory architecture should you specify?
- Your multi-agent design includes (1) a cheap high-volume intent classifier, (2) a reasoning-heavy planning orchestrator, and (3) a code-generation specialist. How should you match tasks to model family capacities?
- While specifying technology components for a Zero Trust multi-agent solution in a regulated industry, which combination best addresses per-agent identity scoping and lateral movement prevention?
- Architects must specify observability so on-call engineers can reconstruct why five agents produced a wrong refund decision. Which observability components should be included?
- A public-sector multi-agent assistant will recommend benefit options. Leadership requires human-AI experience (HAX) and responsible AI controls in the logical design—not only accuracy metrics. Which design package best addresses this?
- You must specify compute for a bursty multi-agent workload: low traffic overnight, sharp peaks during business hours, strict per-tenant isolation goals, and cost sensitivity. Which compute approach best fits?
- A multi-agent engineering team wants consistent local-to-cloud SDLC: reproducible dev environments, IDE support, CLI automation, dependency control, and shared AI coding instructions. Which technology-component specification best supports this?
Develop multi-agent solutions in Azure (30–35%) · 102 questions
- You are implementing advanced prompt engineering for a multi-agent procurement solution in Azure. Supplier agents and a compliance agent share context. Red teaming found agents can be tricked into ignoring policy when a user message says “ignore previous instructions.” Which prompt engineering approach best hardens the system?
- A multi-agent investigation workflow in Microsoft Agent Framework accumulates tool outputs and intermediate reasoning until agents hit context window limits and start dropping critical entity names. Which implementation strategy best manages context across agents?
- You are implementing multi-agent RAG for a legal research suite. A retrieval agent fetches passages; a drafting agent writes memos; a cite-check agent validates sources. Lawyers report fluent memos that miss controlling authority. Which RAG implementation focus best improves multi-agent consumption quality?
- A team is building tool access for multi-agent solutions using Model Context Protocol. They want serverless scale, managed identity to backend APIs, and centralized governance. Which implementation approach best fits Azure-centric MCP tool hosting?
- Your multi-agent travel system calls a third-party inventory API that intermittently returns 503s and occasionally returns schema-valid JSON with nonsensical prices. Which tool integration design best improves resilience and quality?
- Tailwind Traders is implementing multi-agent orchestration for order troubleshooting. A central coordinator must assign work to inventory, shipping, and billing specialist agents and merge results for the customer. Which orchestration pattern should you implement?
- You implement a multi-agent market-research workflow. Competitor pricing scrape, social sentiment analysis, and regulatory news retrieval are independent. A synthesis agent must wait for all three. Which orchestration approach best optimizes task duration?
- You are coding human-in-the-loop for a multi-agent refunds system on Azure. Refunds over $500 require manager approval. Which implementation best matches production HITL patterns?
- A multi-agent HR FAQ system repeatedly answers near-identical policy questions, driving high token cost. You must design caching without serving stale policy after updates. Which caching strategy is most appropriate?
- Your company already runs a mature inventory agent. A new multi-agent fulfillment solution must use it without rewriting it, while also attaching new MCP tools for carrier APIs. Which integration approach is best?
- A Python team on Azure must implement complex multi-agent control flow with branches, loops, and human approval nodes for a claims workflow. They are selecting frameworks aligned to AI-500 skills. Which approach is appropriate?
- You need reusable cross-cutting capabilities across many agents: correlation logging, authorization checks before tool calls, and normalized exception handling. Which design best implements this?
- During a flash sale, your multi-agent support system tries to spawn a new specialist agent per concurrent user message, exhausting model rate limits and budget. Which implementation control is most appropriate?
- You are building an Azure multi-agent assistant that must call internal APIs for order status and, when needed, dynamically select optional tools for returns or exchanges. Which tool-use implementation is most appropriate?
- A support multi-agent system underperforms on company-specific resolution macros. Leaders ask about fine-tuning. Which strategy best aligns with multi-agent development guidance?
- You must integrate enterprise knowledge for multi-agent consumption: product manuals in search indexes, policy packs via MCP-available sources, and semantic retrieval for fuzzy engineer questions. Which design best implements knowledge integration?
- In a multi-agent design review system, architecture, security, and performance agents must critique each other’s findings without a permanent central boss for every message. The workflow still needs eventual aggregation for a final report. Which orchestration choice fits this collaboration style while remaining implementable?
- You are implementing specialist agents in Microsoft Foundry for a claims-processing multi-agent system. The extraction agent sometimes invents policy numbers when source text is incomplete. You need advanced prompt engineering that improves consistency without fine-tuning. Which approach best fits production multi-agent practice?
- A support multi-agent solution uses a triage agent, a billing specialist, and a technical specialist. After a user corrects an earlier account number, the technical specialist continues troubleshooting the wrong account because it only saw the first message. What should you implement so agents behave correctly with evolving conversation context?
- Your legal review multi-agent workflow shows consistent citation formatting errors that few-shot prompting only partially fixes. Leadership asks for a fine-tuning plan for the citation agent’s model. What strategy best matches multi-agent fine-tuning guidance?
- A long-running investigation multi-agent session accumulates tool traces and intermediate reasoning until later agents hit context-window limits and drop earlier evidence. You must keep cross-agent continuity without unbounded growth. Which design should you implement?
- You design memory for a multi-tenant multi-agent assistant: short-term session scratchpads, shared team state for a case, and long-term semantic memory of resolved patterns. Compliance requires tenant isolation and retention limits. Which implementation choice is most appropriate?
- Three agents share a knowledge base: policy lookup, exception handling, and customer explanation. Retrieval often returns irrelevant partial tables, so the exception agent applies the wrong rule. How should you improve the multi-agent RAG architecture?
- You must expose an internal procedures API and a semantic search index to multiple agents without hard-coding HTTP clients in every agent. The platform standard is Model Context Protocol (MCP). What design best integrates knowledge for multi-agent consumption?
- A planner agent must choose among dozens of operational tools depending on the incident type. Hard-coding a fixed tool list in every prompt is unmaintainable. Which approach best builds the tool ecosystem?
- Your team will publish a read-only inventory lookup tool to many agents. Security wants centralized auth, throttling, and versioning. Which implementation aligns with MCP server design on Azure?
- During a parallel multi-agent research run, one MCP tool returns intermittent 429 and timeout errors. Without handling, the whole workflow fails. What should you implement?
- A pricing agent occasionally forwards malformed tool payloads (missing currency, negative quantities) into the quote composer agent, causing bad customer quotes. What control should you add in the tool ecosystem?
- You need a multi-agent claims system where a coordinator assigns work to fraud, coverage, and settlement specialists, aggregates results, and returns one decision package. Specialists should not freely call each other. Which orchestration pattern should you implement?
- An underwriting multi-agent workflow must run credit, property, and compliance checks that do not depend on each other’s outputs, then merge results. Latency SLOs are tight. Which orchestration approach is most appropriate?
- A finance multi-agent system can draft wire instructions but policy forbids sending wires above a threshold without human approval. How should you implement human-in-the-loop?
- Your multi-agent FAQ solution repeatedly pays full inference cost for near-identical policy questions. You need a caching strategy that preserves correctness when policies change. Which design is best?
- Two business units already run separate agents: one on Microsoft Agent Framework and another exposed as an MCP tool server. You must compose them into one multi-agent solution without rewriting both from scratch. What is the most appropriate integration approach?
- Multiple specialist agents in LangGraph / Agent Framework need consistent correlation logging, authorization checks before tool calls, and uniform exception shaping. Duplicating this in every node is error-prone. What should you design?
- A multi-agent claims system uses separate prompts for intake, investigation, and settlement agents. Prompt changes are currently pasted into the portal without versioning, and jailbreak regressions appear after hotfixes. Which implementation approach best improves prompt engineering practice?
- During a long multi-agent troubleshooting session, the orchestrator’s context grows with tool traces and subagent reports until calls fail with context-length errors. You need continuity of key entities without sending the entire history every turn. What should you implement?
- You are implementing knowledge access for a multi-agent product-support solution. A retrieval agent must supply precise passages to a resolution agent and a citation agent. Documents are long, mixed-quality PDFs. Which RAG implementation focus is most important?
- Your team must expose internal order-status and inventory operations to several Foundry agents through a standard tool protocol. Operations should run as scalable Azure compute with existing API backends. Which approach best builds the tool ecosystem?
- A travel multi-agent app calls a flaky third-party fare API. Intermittent 429 and 503 responses currently cause the planner to invent prices. You must harden tool use. What should you implement?
- You need a multi-agent pattern where a central orchestrator interprets user goals, delegates to specialist agents (billing, network, devices), and synthesizes a single response for an ISP support desk. Specialists should not freely chat with each other. Which orchestration pattern should you implement?
- A multi-agent finance workflow can draft vendor payments but must not execute transfers above a threshold without controller approval. Edge cases include missing invoices and conflicting vendor bank details. How should you implement human-in-the-loop?
- Your multi-agent assistant repeatedly re-embeds the same large policy preamble, re-asks identical FAQ questions, and re-calls a slow product catalog tool with the same SKU list within a short TTL. Latency and token costs are high. Which caching strategy should you implement?
- You are implementing a multi-step research workflow on Azure that needs explicit graph control: conditional branches, cycles for refinement, and checkpointed state between nodes. The team is proficient in Python and open-source agent stacks used with Microsoft Foundry model deployments. Which approach best matches the orchestration requirement?
- Multiple specialist agents and tools in your Azure multi-agent solution need the same cross-cutting behaviors: structured logging with correlation IDs, authorization checks before tool execution, and consistent exception handling. Duplicating this logic in every agent is causing drift. What should you implement?
- A multi-agent claims workflow in Microsoft Foundry uses a Planner agent that decides which tools Specialist agents may call. Security review finds that carefully crafted user messages can persuade the Planner to authorize write tools outside the claim’s allowed systems. You must harden prompt design without removing legitimate tool use. What should you implement first?
- An orchestrator agent hands work to three specialist agents over a long-running case. Each specialist appends full intermediate reasoning and raw tool dumps into a shared conversation buffer. After several hops, agents begin dropping earlier constraints and contradicting prior decisions. Which approach best implements context management between agents?
- You design a multi-agent RAG solution where a Research agent retrieves policy documents and a Writer agent drafts customer replies. The Research agent often returns partially relevant chunks, so the Writer invents policy details. Product owners require higher retrieval precision without forcing the Writer to search itself. What should you implement?
- Your team must expose an internal order-status API to several Foundry agents as standardized tools that multiple orchestrations can reuse. Agents should discover tools consistently, and platform engineers want centralized auth, throttling, and versioning at the edge. Which design best fits Microsoft multi-agent tool ecosystems?
- A Fulfilment agent calls a warehouse inventory tool. Intermittent 503 responses and occasional schema-mismatched payloads cause the agent to invent stock levels and continue the workflow. You must implement robust tool integration behavior. What combination should you design?
- A customer onboarding workflow must (1) run KYC document checks, (2) provision cloud resources, and (3) send a welcome sequence. Steps 1 and 2 are independent and can run concurrently; step 3 must wait for both to succeed. Compliance requires a single coordinating agent to assign work and merge outcomes. Which orchestration pattern should you implement?
- A finance multi-agent system can draft vendor payments. Policy requires that any payment instruction above a threshold receive human approval before the payment tool executes. Agents must not bypass this for “urgency.” How should you implement human-in-the-loop?
- You are implementing prompt strategy for a multi-agent procurement system in Azure. Specialist agents receive dynamic context (vendor policies, cart state) and must resist prompt injection from untrusted supplier portal text. Which approach best implements advanced prompt engineering for this solution?
- In a multi-agent research workflow on Azure, the Researcher agent accumulates long web-tool traces that must be handed to a Writer agent without exceeding context limits or losing key citations. Which context management design should you implement?
- You are implementing multi-agent RAG for a legal discovery suite. A Searcher agent retrieves passages; an Analyst agent reasons over them. Early tests show poor retrieval precision and lost clause boundaries. Which RAG design change best addresses embedding quality and retrieval precision?
- Your team must expose an internal pricing calculator to multiple agents as a governed tool. Developers prefer a lightweight serverless implementation that can be reached through Model Context Protocol clients. Which implementation approach best fits Azure multi-agent tool ecosystems?
- A travel multi-agent solution calls an airline inventory tool. Intermittent 503s and schema validation failures occur. Which tool error-handling and fallback design should you implement?
- You implement orchestration for a due-diligence multi-agent workflow: financial analysis, legal review, and market research can run independently, then a partner agent synthesizes findings. Which orchestration pattern should you implement?
- A banking multi-agent system can propose wire transfers. Policy requires human approval for wires over $10,000, supports manager override on edge cases, and must record who approved what. How should you implement human-in-the-loop?
- A multi-agent support system repeats similar policy questions and identical FAQ answers thousands of times daily. Token spend is high. Which caching strategy design best fits multi-agent optimization on Azure?
- During month-end, a multi-agent finance close system must process thousands of entity reconciliations. Naively spawning unlimited agent instances exhausts model rate limits. What should you implement?
- Contoso acquired a partner that already runs a warranty specialist agent. Your orchestrator is on Microsoft Foundry. You must integrate the external agent without sharing long-lived secrets in prompts and without reimplementing warranty logic. Which approach is best?
- Your team prefers graph-based control flow with explicit nodes, edges, and conditional branching for a multi-agent claims pipeline on Azure. Microsoft Agent Framework is available, but the engineers already prototyped complex cycles in LangGraph. Which implementation choice best matches the requirement for advanced graph orchestration?
- You need reusable cross-cutting capabilities across many agents: structured logging, authorization checks before tool calls, and normalized exception handling. Where should these capabilities live in the multi-agent solution design?
- An agent receives JSON from a customs tariff tool. Occasionally fields are missing or units are wrong, which previously caused incorrect duty calculations. What should you implement before the agent commits downstream actions?
- You are integrating knowledge for multi-agent consumption: product manuals in Azure AI Search, live inventory via an MCP tool, and policy PDFs requiring semantic retrieval. Which knowledge integration design is most appropriate?
- A support multi-agent system struggles with company-specific shorthand that few-shot prompting only partially fixes. Leadership asks about fine-tuning. What strategy should you design?
- You are coding a customer-service multi-agent solution where a single user-facing orchestrator must delegate to billing, shipping, and returns specialists without those specialists talking directly to end users. Which pattern should you implement?
- You implement multi-agent memory for a public-sector casework system. Requirements include encryption, retention limits, session isolation between citizens, and purge on case closure. Which memory implementation approach is correct?
- A multi-agent claims workflow in Microsoft Foundry keeps a shared working context that accumulates every agent’s intermediate reasoning, tool payloads, and retrieved passages. After several long-running cases, later agents begin truncating mid-sentence because the context window is full, and critical claim identifiers from early turns are dropped first. Which approach best restores usable shared context without losing high-value entities?
- A platform team wants specialist agents in Microsoft Foundry to discover and call a shared set of internal operations (inventory lookup, price quote, order status) through a standardized protocol rather than each agent embedding its own HTTP client and schema. Operations already run as Azure Functions behind Azure API Management. Which design best exposes these operations for multi-agent consumption?
- A market-research multi-agent system must, for each product brief, simultaneously gather competitive pricing, social sentiment, and regulatory notes from three independent specialist agents, then produce one consolidated research pack. Latency budgets require that the three specialists run at the same time rather than waiting on each other. Which orchestration pattern should the team implement?
- An enterprise FAQ multi-agent solution on Azure repeatedly pays for full model completions on near-identical user questions such as “How do I reset my VPN password?” and “What’s the process to reset VPN credentials?” Specialists and the orchestrator share the same grounding corpus. The team wants to cut token cost and latency for semantically equivalent questions while still allowing true novel questions to hit the models. Which caching strategy best fits?
- A procurement agent calls an external supplier API tool that occasionally returns partial JSON (missing required fields), HTTP 200 with an error object in the body, or numeric fields encoded as strings. The agent has begun treating these payloads as successful quotes and recommending invalid purchase orders. What is the most direct remediation within the tool ecosystem design?
- A customer-service multi-agent system can draft refund decisions, but company policy requires that any refund over $500 must be approved by a human supervisor before the refund tool executes. Agents may still gather facts and draft the recommendation autonomously below that threshold. Which implementation best meets the requirement?
- An operations agent has access to dozens of tools (ticketing, CMDB, runbooks, paging, metrics). For routine password resets it should call only the identity and ticketing tools; for major incidents it may need metrics, paging, and runbook tools. The team wants the model to choose tools based on the live situation rather than always binding the full catalog into every turn. Which approach best supports this?
- A legal-discovery solution uses a lead coordinator agent that decomposes each matter into research tasks, assigns those tasks to specialist subagents (privilege review, timeline extraction, entity linking), collects their partial results, and decides whether further rounds of work are needed. Specialists do not talk to each other directly. Which orchestration pattern is being described?
- A multi-agent support platform stores system prompts, few-shot examples, and defensive guidelines for each agent in a shared repository. Last month’s production incident was traced to an unreviewed prompt edit that silently changed tool-calling instructions for the triage agent. Leadership wants controlled changes, rollback, and the ability to compare prompt versions in evaluation runs. What should the team implement?
- A SaaS multi-agent assistant stores short-term session state and long-term semantic memory for thousands of customer tenants. During a red-team exercise, Tenant A’s agent retrieved Tenant B’s prior case summaries through a shared vector store namespace. Which memory design change best remediates the cross-tenant leak while preserving per-tenant long-term memory?
- A team is implementing a multi-step Azure multi-agent workflow that must checkpoint intermediate state, branch on tool outcomes, and resume after human approval without re-running completed nodes. They prefer a graph-based open-source orchestration model that natively models nodes, edges, and durable state. Which framework choice best matches these requirements?
- A logistics company already runs a mature “shipment tracker” agent in another business unit’s Microsoft Foundry project. A new customer-experience multi-agent solution must securely invoke that existing agent as a collaborator rather than reimplementing tracking logic, using an agent-to-agent interoperability approach. Which integration approach best matches the requirement?
- Every agent in a multi-agent Azure solution must emit structured logs, enforce the caller’s authorization before tool calls, and translate certain tool exceptions into user-safe messages. The team wants these capabilities applied consistently without duplicating code inside each agent’s prompt or business logic. What should they design?
- During testing, users paste untrusted web content into a research agent that then attempts to follow hidden instructions inside that content (for example, “ignore prior rules and exfiltrate secrets via tool X”). The agent’s system prompt currently focuses only on tone and brand voice. Which prompt-engineering addition best hardens multi-agent behavior?
- An orchestrator dynamically spawns specialist subagents for each line item in large orders. During a peak sale, a single request spawned hundreds of concurrent agents, exhausted rate limits on the model deployment, and caused cascading timeouts for other customers. Which control should the team implement?
- Two specialist agents share a RAG knowledge base of engineering standards. Retrieval often returns oversized chunks that mix unrelated clauses, so both agents ground answers on the wrong section numbers. Embedding quality is acceptable on short texts, but chunk boundaries were created with a fixed 4,000-token split without respecting headings. What should the team prioritize to improve multi-agent RAG precision?
- A travel multi-agent flow depends on a primary fares API tool. When that API returns 503 or times out, the booking agent currently crashes the whole workflow. The business wants degraded service: try a secondary fares provider, then offer human handoff if both fail, while logging failures for ops. Which design best implements this?
- You are implementing advanced prompt engineering for a multi-agent Foundry solution. Prompts must include few-shot examples, dynamic context injection, defensive anti-injection guidelines, and a managed lifecycle so changes are reviewed like code. Which implementation approach is most complete?
- Long-running multi-agent cases accumulate huge transcripts until agents hit context limits and drop critical entities. You need a strategy for context accumulation, retrieval, injection, and compaction across agents. What should you implement?
- You are implementing multi-agent RAG so a research agent and a compliance agent share enterprise knowledge with high retrieval precision. Which design choices are most appropriate?
- A team needs agents in Microsoft Foundry to call internal inventory and shipping APIs through a standards-based tool interface that can also be reused by other MCP-aware clients. Which implementation best fits?
- A multi-agent workflow calls an external pricing API tool that intermittently returns 503s. You need robust tool error handling and fallback so the orchestration does not crash the user session. What should you implement?
- You are implementing multi-agent orchestration on Azure using a graph-based framework. The design needs an orchestrator that delegates to specialist subagents, supports conditional branches, and can wait for human approval on high-risk steps. Which approach fits?
- Token costs for a multi-agent support system are dominated by repeated similar questions and repeated system prompts. Which caching strategy should you implement?
- Your company already has a standalone document-processing agent and a separate MCP-based records tool server. A new multi-agent solution must reuse both without rewriting them from scratch, while keeping secure integration. What should you do?
- An orchestrator must analyze 40 independent contract clauses. Sequential specialist calls exceed the latency SLO. How should you implement scaling of multi-agent execution?
- Multiple specialist agents need consistent request logging, authorization checks, and exception handling without duplicating boilerplate in every agent. What should you implement?
- You are implementing a multi-agent claims-processing workflow in Microsoft Foundry. Subagents receive instructions assembled from shared system prompts plus per-task user context. Security review finds that a malicious user message can override the underwriting agent's refusal rules by injecting "ignore previous instructions." What should you implement first in the prompt design?
- A research orchestrator and three specialist agents collaborate on long investigations. After about twenty turns, specialists begin contradicting earlier findings because the shared working context is truncated mid-entity. You need a context-management approach that preserves entity continuity across agents without exceeding model context limits. What should you implement?
- Contoso is building a multi-agent operations desk. Agents must call internal inventory, ticketing, and change-management APIs through a consistent tool interface with centralized auth and observability. Leadership wants standards-based agent-to-tool integration rather than ad-hoc function schemas per framework. Which approach best fits Microsoft Foundry multi-agent tooling guidance?
- A support solution needs a coordinator agent that classifies intent, then routes work to billing, technical, or shipping specialist agents. Specialists should not call each other directly; only the coordinator consolidates results for the user. Which multi-agent orchestration pattern should you implement?
- Your multi-agent procurement workflow can draft purchase orders and call a create-PO tool. Finance requires that any PO above $10,000 pause for human approval before the tool runs, with the ability to override or reject. How should you implement this control?
- An internal knowledge multi-agent system answers similar HR policy questions thousands of times per day. Token spend is high even though many user phrasings map to the same grounded answer. You need lower latency and cost without serving stale policy after documents change. Which caching strategy is most appropriate?
- Two business units already run production agents—one on Microsoft Agent Framework in Foundry and one on an external LangGraph service. You must compose them into a single customer journey without rewriting both codebases or sharing long-lived static secrets in prompts. What is the best integration approach?
Evaluate, optimize, and monitor multi-agent solutions (20–25%) · 72 questions
- After deploying a multi-agent customer-email system in Microsoft Foundry, quality fluctuates weekly. You need continuous improvement beyond occasional manual spot checks. Which evaluation approach best fits?
- Operators notice a long-running multi-agent investigation “forgets” earlier entities: names morph, amounts change, and later agents contradict earlier tool results after repeated summarization. Which diagnosis and response is most appropriate?
- Finance flags rising spend on a multi-agent Azure solution. Engineering must monitor reliability and cost: token burn, loop storms, tool-call fan-out, and workflow failures—with actionable alerts. Which strategy best addresses this?
- A product team runs a multi-agent claims workflow in Microsoft Foundry. Human spot-checks cannot keep up with nightly regression volume. They need an automated evaluation loop that scores agent final answers for faithfulness to retrieved knowledge and for policy compliance, then flags regressions before promotion. Which approach best meets this continuous-improvement need?
- After a production incident, engineers must reconstruct why a supervisor agent approved an incorrect subagent tool call. Logs exist in several microservices, but they cannot link one user session across the orchestrator, tool gateway, and model calls. Which observability design should they implement in Microsoft Foundry?
- Users report that a long-running research multi-agent loses early constraints (budget cap and excluded vendors) after many tool turns, then invents conflicting recommendations. Token usage per session is near the model limit. Which diagnosis best explains the failure mode?
- A parallel research swarm repeatedly re-queries the same MCP search tool in tight loops, driving token spend and latency far above SLA. Product wants lower cost without removing the tool entirely. Which optimization strategy should you implement first?
- Over six weeks, a support multi-agent’s automated evaluation suite shows rising policy-violation rates even though code deploys were frozen. Knowledge articles were updated daily by content owners. What monitoring strategy best detects and explains this pattern?
- Finance requires chargebacks for a shared multi-agent platform used by three business units. Today all Foundry usage lands under one subscription with no attribution. What cost-management design should the platform team implement?
- Before promoting a multi-agent mortgage underwriting workflow, compliance requires human review of a statistically valid sample of agent decisions, including tool inputs and final rationales. How should the team design this process in Microsoft Foundry?
- An orchestrator fans out ten research subagents sequentially even though their tasks are independent. End-to-end duration exceeds the five-minute SLA, while the model deployment still has headroom before rate limits. What should you do to optimize task duration?
- A platform squad wants evaluations that isolate failures across memory, knowledge retrieval, tools, and prompts rather than only scoring final answers. Which evaluation design best supports root-cause localization?
- Before promoting a multi-agent workflow in Microsoft Foundry, product owners must score groundedness and task success on a labeled set. Automated metrics alone are not trusted yet. What should you design?
- After a release, users report the research agent forgets case facts mid-session, tools return unused, and prompts drift from the approved template. What evaluation strategy should you implement?
- Your multi-agent batch job finishes too slowly. Profiling shows sequential specialist calls and frequent 429s from the model endpoint. What optimization should you apply first?
- Analysts notice the investigation agent forgets entities introduced early, summaries slowly change key facts, and vector recall sometimes returns topically similar but wrong cases. Which diagnosis best matches these context-window issues?
- You need continuous improvement for a multi-agent support solution: scalable scoring of free-text answers, periodic synthetic hard cases, and incorporation of user thumbs-down. Which approach best fits?
- Operations needs a reliability strategy for multi-agent production: detect unhealthy agents, workflow failures, quality regressions, and behavioral drift, then trigger remediation. What should you design?
- A multi-agent concierge promises 99.9% workflow success within 8 seconds end-to-end. Leadership wants platform and solution monitoring aligned to that SLA. What strategy is appropriate?
- Token spend for a planner-heavy multi-agent graph spiked after agents began long tool-call loops. What optimization strategy should you implement?
- Finance requires showback for multi-agent usage across business units, with quotas and alerts before overage. What monitoring and management approach should you design?
- Debugging a failed multi-agent run is hard because prompts, token counts, and tool calls are scattered without a single timeline. What should you implement in Microsoft Foundry?
- After a prompt change to the triage agent, automated offline scores look stable, but live users report more misroutes. What monitoring gap should you close?
- A rare multi-agent failure occurred once in production. Engineers cannot reproduce it from logs that only store final answers. Which observability capability is most important to add?
- You need a repeatable way for domain experts to score multi-agent runs in Microsoft Foundry before promoting a new orchestrator version. Automated metrics exist but miss subtle policy tone issues. What should you design?
- After a release, overall task success drops. You suspect either degraded memory retrieval, a broken tool contract, or a prompt regression—but a single end-to-end pass/fail metric cannot tell you which. What evaluation approach should you implement?
- A parallel research workflow spawns eight specialist agents that all hit the same model deployment and the same external search API. Median task duration has doubled; logs show frequent 429 responses. What optimization should you apply first?
- A long-running multi-agent investigation repeatedly summarizes its own history to fit the context window. Over time, critical entity attributes change incorrectly in the summary (for example, customer region and contract tier), and later agents act on the wrong facts. Which diagnosis and remediation pair is best?
- You want continuous improvement for a multi-agent support solution: large volumes of traces, limited expert time, and a need to catch semantic regressions between weekly releases. Which approach best fits?
- Operations reports that a multi-agent workflow’s quality slowly worsened over two weeks without a code deploy. Tool schemas are unchanged. You need a monitoring strategy that catches this class of failure and supports response. What should you design?
- A multi-agent booking solution has a contractual SLA: 99.9% successful completion of the booking workflow within 45 seconds for interactive sessions. Platform CPU is healthy, but customers report timeouts when the orchestrator waits on a slow third-party MCP tool. What monitoring design addresses SLA adherence?
- Cost and latency spiked after a prompt change. Traces show agents repeatedly calling the same tool and re-planning in long loops until max tokens. Which optimization strategy should you implement?
- A shared multi-agent Foundry platform serves three business units. Finance demands showback of model and tool spend per unit; one unit is exhausting the shared quota and starving others. What cost-management strategy should you design?
- Debugging a failed multi-agent run requires reconstructing which orchestrator decision led to which tool call and model response across services. Today logs are disconnected per component. What should you implement in Foundry-oriented observability?
- Interactive multi-agent sessions truncate oldest turns when the context window fills. Users report that early constraints (“never email the customer directly”) disappear mid-session and agents violate them later. Which remediation best targets sliding-window amnesia?
- A team updates system prompts for three specialist agents weekly. Last month a prompt edit improved tone but increased incorrect tool arguments. You need an evaluation practice that prevents similar regressions. What should you do?
- Your team must evaluate a multi-agent claims solution in Microsoft Foundry before wider release. Stakeholders want structured human review of agent outcomes, not only automated scores. Which approach best designs the human review process?
- Users report that a long-running multi-agent project assistant forgets earlier constraints after many turns. Engineers notice summaries gradually omit the original budget cap while vector recall returns loosely related notes. Which diagnosis and remediation pairing is best?
- After launch, you need continuous improvement for a multi-agent assistant: detect quality regressions, generate hard test cases, and fold user thumbs-down signals into optimization. Which strategy best fits AI-500 continuous improvement practices?
- A multi-agent claims assistant in Microsoft Foundry keeps a rolling conversation context using a fixed-size sliding window. Over long claims investigations, the policy-number entity and claimant name that were established in the first few turns disappear from later agent prompts, and downstream agents begin inventing identifiers. Offline evaluation shows retrieval still finds the correct claim, but the writing agent no longer mentions the original entities. Which diagnosis best explains the failure mode, and what should the team address first?
- A product team runs a multi-agent research-and-draft pipeline in Microsoft Foundry. They want a continuous improvement loop that scores draft quality on faithfulness and structure at scale, then feeds weak cases into prompt and tool refinements. Human labels exist only for a small golden set. Which approach best meets these goals?
- Tracing in a Foundry multi-agent support solution shows that a troubleshooting agent sometimes issues more than forty tool calls for a single ticket, repeatedly querying similar telemetry endpoints with slightly rephrased parameters. Token spend and latency spike on those tickets even when a useful answer was available after five or six calls. Which optimization best reduces waste while preserving legitimate multi-step diagnosis?
- Contoso runs three business units on a shared multi-agent Foundry platform. Finance complains that monthly Azure OpenAI spend is opaque: one unit's research swarm appears to consume most tokens, but chargebacks are still split evenly. Platform engineering must improve cost accountability without redesigning every agent. What should they implement first?
- A multi-agent project team shares a rolling conversation summary that each specialist agent updates after its turn. Over multi-day projects, the summary gradually invents milestone dates and drops constraints the user stated early on. Individual agent traces look locally coherent. Which evaluation focus best detects this class of failure?
- After a failed multi-agent checkout flow, support engineers can see each microservice log stream but cannot reconstruct which orchestrator decision led which subagent to call which payment tool for that one customer order. What observability design should they implement in Foundry and related services?
- Six weeks after launch, a multi-agent HR policy assistant still passes its original pre-release evaluation suite, but employees report that answers have become more verbose, less actionable, and occasionally cite retired policy versions after knowledge-base updates. Which monitoring strategy best closes this gap?
- An architect plans to cut median latency of a six-agent research workflow by running independent retrieval specialists in parallel. During a pilot, total wall-clock time barely improves and 429 rate-limit errors surge against a shared Azure OpenAI deployment and a shared search service. What should the team do to optimize task duration realistically?
- A regulated insurer must approve any change to a multi-agent underwriting assistant before production. Automated Foundry evaluations score the suite green, but risk stakeholders require evidence that humans reviewed edge cases the metrics might miss. Which process design best satisfies both speed and governance?
- A multi-agent claims team needs continuous quality scoring on thousands of nightly agent runs. Human review can cover only a small sample. They want automated semantic scoring of final answers against policies, with periodic human calibration, feeding a semantic optimization loop. Which approach best fits Foundry-oriented continuous improvement?
- Support engineers debugging a multi-agent failure in Microsoft Foundry can see that a user request failed, but they cannot reconstruct which agent ran, which tools were called, how many tokens were consumed, or how child spans related to the parent request. Which observability implementation should the team prioritize?
- Users report that a long-running multi-agent planning session “forgets” goals stated at the beginning of the conversation after many turns, even though no explicit memory wipe was coded. Engineers confirm the implementation uses a simple last-N message window with no summarization of older turns. Which context-window issue is this, and what is the appropriate diagnostic framing?
- Cost dashboards show one multi-agent workflow consuming extreme token volume because the orchestrator and a critic agent refine the same paragraph for dozens of rounds with no stop condition, each round including full history and multiple tool calls. Which optimization strategy best targets this failure mode?
- After a prompt change to the triage agent, automated offline scores looked fine on a tiny smoke set, but production customer satisfaction for multi-agent resolutions dropped 12% week over week. On-call had no alert because only infrastructure uptime was monitored. What monitoring capability should be added?
- A team must expand evaluation coverage for a multi-agent benefits advisor before open enrollment. Real labeled production transcripts are scarce because of privacy constraints. They need diverse scenarios (edge cases, adversarial phrasings, multi-turn tool paths) to test prompts, tools, and memory. Which continuous-improvement technique best addresses the gap?
- Finance reports that one business unit’s multi-agent experiment consumed the majority of the shared Azure OpenAI quota, starving production agents in other units. Leadership wants usage visibility by team, hard allocations, and chargeback. Which monitoring and management strategy addresses this?
- A multi-agent research workflow currently runs four independent specialist agents one after another; total duration is the sum of all four. Model deployment rate limits still have headroom for higher concurrency. Product requires lower p95 latency without changing answer quality targets. What optimization should the team apply first?
- Before promoting a new multi-agent configuration in Microsoft Foundry, the responsible AI lead requires structured human review of sampled traces: correctness, tool appropriateness, and policy adherence—not only automated scores. Which evaluation practice should the team implement?
- A multi-agent session uses recursive summarization to fit long histories into context. After many cycles, the running summary still “sounds coherent” but has silently changed a key constraint from “exclude vendor X” to “prefer vendor X,” and downstream agents act on the wrong constraint. Which context issue is this, and what is the right response?
- Platform ops needs a reliability strategy for a production multi-agent system: detect when individual agents are unhealthy, when workflows fail mid-orchestration, when traces cannot be correlated, and when automated remediation should page on-call or restart a failed step. Which monitoring strategy scope is most complete?
- A multi-agent release checklist currently only unit-tests Python helpers. Failures in production are often caused by bad tool schemas, weak retrieval, prompt regressions, or memory mis-scoping—not helper functions. What evaluation design should the team add before promotion?
- Before promoting a multi-agent customer-support graph from staging to production in Microsoft Foundry, the quality team must score groundedness, tool-choice appropriateness, and tone on a fixed set of scenarios with human reviewers. What should you design first?
- Users report that a long-running multi-agent planning session “forgets” constraints agreed in the first ten minutes once the transcript grows large, even though no code bug resets state. Traces show oldest turns disappearing from the prompt assembly. Which context-window issue is this, and what is the best remediation direction?
- You need a continuous improvement loop for a multi-agent sales assistant. Human review capacity is limited, but you must detect quality regressions in grounded product claims weekly. Which approach aligns with AI-500 continuous improvement practices?
- A multi-agent research workflow fans out five specialist agents that each call the same Foundry model deployment. End-to-end duration is poor: some specialists idle while others hit HTTP 429s. What optimization strategy should you apply first?
- After enabling automatic conversation summarization between agent handoffs, auditors find that a critical contractual exclusion present in the original transcript is missing from later specialist contexts, replaced by a softer paraphrase that changes meaning. Which issue is occurring and what should you do?
- Incidents in production show incomplete explanations: Azure Monitor has some dependency logs, but engineers cannot reconstruct which orchestrator decision led which subagent to call which tool for a single user session. What observability design is required?
- A peer-to-peer multi-agent debate pattern occasionally enters long tool-call loops, driving token spend 20× above budget for a minority of sessions. You need controls that protect cost without killing legitimate multi-step work. What should you implement?
- Over three weeks, automated evaluations show a gradual drop in tool-selection accuracy after several prompt and model-version changes, though infrastructure uptime remains 99.9%. Which monitoring concern is this, and what should you add?
- Finance requires each business unit to pay for its multi-agent Foundry consumption. Today all agents share one deployment with no attribution, so chargebacks are impossible. What monitoring and management strategy should you design?
- You are building an evaluation suite for a multi-agent RAG helper that also calls MCP tools. Stakeholders want coverage beyond final answer text. Which evaluation design best matches the exam’s multi-agent validation scope?
- Platform engineers must alert when any multi-agent workflow exceeds a token budget or repeatedly fails a critical tool stage. Which Foundry-oriented tracing implementation details are essential?
- You want to stress-test a multi-agent assistant for rare but high-impact failures (prompt injection, wrong tool under urgency, contradictory memory) without waiting months for organic traffic. Which continuous improvement technique should you prioritize?
Secure, govern, and deploy multi-agent solutions (20–25%) · 66 questions
- Multiple agents need database credentials and third-party API secrets. Developers currently paste secrets into agent environment variables in plain text and rarely rotate them. Which approach meets security requirements for multi-agent solutions on Azure?
- A bank is preparing to launch a multi-agent customer assistant. Security wants shift-left adversarial testing of prompt injection, tool abuse, and jailbreak attempts before production. Which Foundry-aligned practice should they adopt?
- An enterprise multi-agent system must block toxic user inputs, prevent disallowed tool calls, validate tool responses before they re-enter the model, and filter unsafe final answers. Security rejects a single check only on the final response. What guardrail strategy should you design?
- A multi-agent expense system must call a user’s personal calendar and expense APIs using that user’s permissions—not a superuser service account—so agents cannot read other employees’ data. Which authentication approach is most appropriate?
- Platform engineering must release a new orchestrator version that changes tool-routing logic. They want limited blast radius, fast rollback, and production traffic validation—not a big-bang cutover. Which release methodology best fits?
- In a hub-and-spoke multi-agent design, the research subagent only needs read access to a knowledge index, while the payment subagent needs to initiate refunds. Security requires preventing lateral movement if the research agent is compromised. What identity design should you apply?
- A team deploys multi-agent infrastructure and prompt/tool packages by clicking in the Azure portal and copying files over SSH. Releases are inconsistent across environments. What CI/CD approach should they adopt?
- A healthcare multi-agent assistant must never output controlled substance dosage advice or attempt to schedule procedures that require clinician approval. Built-in generic content filters are helpful but incomplete for these clinical policies. What should the team implement?
- A multi-agent platform in a regulated tenant must ensure model endpoints, vector stores, and tool backends are not reachable from the public internet, while agents still call them securely from Azure. Which design best matches Zero Trust network boundaries?
- A canary of a new planner prompt set caused a spike in unsafe tool invocations. Automated gates paused expansion. What should the release design specify next for multi-environment multi-agent deployments?
- Security wants proof that new multi-agent guardrails block prompt injection and disallowed tool combinations before production. Production traffic is sparse and cannot supply enough attack variety. Which testing approach should they use?
- You must reduce prompt injection and unsafe tool use in a multi-agent Foundry deployment. Security asks for defense in depth across the agent pipeline. Which guardrail strategy is most appropriate?
- A financial services firm is deploying a multi-agent claims workflow in Microsoft Foundry. Three specialized agents must call different backend APIs: a document-extraction agent, a policy-lookup agent, and a payout-recommendation agent. Security requires that a compromised agent cannot call APIs outside its role. You must enforce least privilege without embedding long-lived secrets in agent configuration. Which approach best meets these requirements?
- You are building a multi-agent assistant that helps employees query their own Microsoft Graph mail and calendar. Downstream tools must only see data the signed-in user is already allowed to access. Agents run as Azure-hosted services and call Graph on the user’s behalf. Which authentication pattern should you implement for the agent-to-Graph calls?
- A multi-agent order-fulfillment system stores third-party shipping API credentials and signing certificates used by tool connectors. Compliance requires centralized secret storage, role-based access to secrets, automated rotation, and encryption at rest. Agents run on Azure with managed identities. Which design best satisfies these requirements?
- Your platform team wants to catch prompt-injection, tool-abuse, and jailbreak risks in a multi-agent Foundry solution before production release. They already run unit and integration tests in CI. Which shift-left security practice best extends that pipeline for agentic threats?
- A multi-agent customer-support system can read knowledge bases, call refund tools, and generate customer-facing replies. Risk review found that harmful content can enter via user chat, malicious tool payloads, or model-generated text. Leadership wants layered controls, not a single filter at the end. Which guardrail strategy best matches multi-intervention design?
- A hospital is piloting multi-agent triage helpers that draft clinical summaries. Beyond generic content safety, policies forbid diagnosing rare conditions without clinician review, forbid prescribing language, and require PHI redaction before any external tool call. How should you implement these requirements?
- You must roll out a major change to a multi-agent orchestration graph (new planner prompt, updated tool schemas, and a new subagent). Leadership wants limited user blast radius, fast rollback, and production traffic validation—not only lab evaluations. Which release methodology best fits?
- Your organization manages multi-agent solutions across Dev, Test, Acceptance, and Production. Agent definitions, MCP tool registrations, Azure resources, and evaluation suites must stay consistent and auditable. Which CI/CD approach best meets multi-environment release requirements?
- A regulated enterprise deploys multi-agent workloads that call Azure OpenAI-compatible model endpoints, Azure AI Search, and internal APIs. Policy forbids public ingress to data-plane services where private connectivity is available. Agents run on Azure Container Apps. Which network design best aligns with resource access and network boundary controls?
- After implementing multi-stage guardrails for a multi-agent procurement assistant, the security team needs evidence that the controls block known attack classes before each release. Production traffic is sparse for rare jailbreaks. What is the best validation approach?
- A multi-agent system previously used one shared service principal for the orchestrator and all specialist agents to call Azure resources and tools. A security assessment flags lateral movement risk if any agent is compromised. What should you implement?
- An assistant agent must call a downstream Microsoft Graph API to read only the signed-in user’s calendar. Compliance forbids the agent from using an app-only permission that can read all calendars in the tenant. Which authentication flow should you design?
- Specialist agents need database credentials and third-party API keys at runtime. Today secrets are environment variables baked at deploy time and rarely rotated. How should you design secrets management?
- Before production, security wants proactive adversarial testing of the multi-agent system for prompt injection, tool abuse, and jailbreaks. You are implementing shift-left security in Microsoft Foundry. Which approach best aligns?
- A regulated multi-agent assistant must block toxic user inputs, prevent disallowed tool calls, sanitize untrusted tool responses, and filter unsafe final answers. A single output filter has been bypassed by malicious tool-return content influencing the model. What guardrail strategy should you implement?
- A healthcare multi-agent system must never recommend changing prescribed medications and must refuse diagnosis claims. Built-in generic safety filters help but do not encode this clinical policy. What should you add?
- You implemented multi-stage guardrails but lack evidence they hold against creative bypasses. QA has limited real incident data. How should you validate guardrails?
- You are releasing a new orchestrator graph that changes tool-routing logic. Full big-bang production cutover is considered too risky. Traffic is sticky per session. Which release methodology best fits?
- Your DTAP pipeline promotes multi-agent configs (prompts, tool allowlists, model deployment references) as versioned artifacts. A production canary shows elevated tool-failure rates after promotion. What rollback design should already be in place?
- A platform team wants CI for a multi-agent repo. Engineers currently only run occasional manual chat trials. Which testing strategy should you specify for integration and release?
- You must deploy Foundry-related infrastructure, agent configurations, and guardrail policies consistently across Dev, Test, Acceptance, and Production. Drift between environments caused a production-only outage last quarter. What CI/CD requirement should you specify?
- A multi-agent expense system must call a line-of-business API that authorizes actions based on the signed-in employee's identity and group memberships. The orchestrator and tools currently use a single long-lived application secret with broad application permissions, so every agent acts with the same superuser power. Which authentication design best aligns with least privilege for user-scoped actions?
- A customer-facing multi-agent concierge can read email, call booking tools, and generate replies. Red-team tests show jailbreak text sometimes reaches tools, tool responses sometimes contain competitor-injected spam, and final replies sometimes leak internal system instructions. Which guardrail strategy best matches a multi-intervention design?
- Several Foundry agents retrieve third-party API keys and certificates at runtime. Today secrets are stored in a shared configuration map baked into container images, and rotation requires rebuilding every agent image. Which approach best meets secrets management requirements for multi-agent solutions on Azure?
- A platform team wants shift-left security for a new multi-agent claims workflow before the first production canary. They already run dependency scanning on infrastructure code. Which additional practice best targets agent-specific abuse paths such as prompt injection and tool misuse?
- A team must ship a risky orchestrator prompt change that alters how three specialist agents are sequenced. Automated evals look good in staging. They need a production release method that limits blast radius if live behavior diverges. Which release methodology best fits?
- In a Zero Trust multi-agent design, a low-privilege FAQ agent was compromised via prompt injection and attempted to invoke the wire-transfer tool bound only to a payments agent. The tool call succeeded because both agents shared one managed identity with contributor rights on all tools. What architectural fix best prevents this lateral movement?
- A pharmaceutical multi-agent system must never recommend off-label drug uses or invent clinical claims. Generic toxicity filters catch insults but miss domain violations. How should the team design guardrails and validation?
- An operations team deploys multi-agent solutions across DTAP environments and needs a production cutover strategy that can reverse a bad agent package within minutes if health probes fail. Which approach best meets that requirement?
- A platform team wants every multi-agent change—prompt files, tool schemas, orchestration graphs, and Azure infrastructure—to follow the same quality bar before production. Which CI/CD requirement set best fits AI-500 deployment guidance?
- A multi-agent system in Azure must call internal finance APIs that must not be reachable from the public internet. Agents run in Foundry-connected compute inside a virtual network. Which network design best enforces this boundary?
- An architect reviews RBAC for a multi-agent solution. The research agent only needs read access to a knowledge index and a web-fetch tool. The fulfillment agent needs to create support tickets. Currently both agents use a role that can read secrets, write tickets, and delete indexes. What change best applies least privilege?
- A regulated multi-agent assistant can read user input, call internal tools, receive tool responses, and produce user-facing answers. Security review found filtering only on the final answer, allowing jailbreak text in inputs and sensitive data in tool responses to influence intermediate reasoning and tool selection. Which guardrail strategy best addresses the gap?
- A financial-services team is deploying a multi-agent solution in Microsoft Foundry. A research agent must read SharePoint knowledge bases, while a settlement agent must call a payment API. Security requires that a compromise of the research agent must not allow access to payment credentials. Which design best enforces this Zero Trust boundary?
- An orchestrator agent in Microsoft Foundry accepts an end-user chat session and must call a downstream specialist agent that then invokes Microsoft Graph using the signed-in user’s permissions. The specialist must never use application-only permissions for mailbox data. Which authentication flow should you implement between the user, orchestrator, and specialist tools?
- A healthcare multi-agent system must block PHI leakage, unsafe tool use, and non-compliant model outputs. Architects want defense-in-depth rather than a single filter at the end of the chat. Which guardrail strategy best matches multi-intervention design for multi-agent solutions?
- Your multi-agent solution stores third-party API keys in Azure Key Vault. Agents currently fail for up to an hour after secrets rotate because they cache values at process start. Operations needs rotation without multi-hour outages and without putting secrets in source control. What should you implement?
- Before promoting a multi-agent Foundry solution from test to production, the security team wants automated adversarial probing of prompts, tool misuse, and jailbreaks as part of CI—not only annual pen tests. Which approach best applies shift-left security for this solution?
- A platform team must release a new orchestrator prompt and tool graph for a customer-facing multi-agent app with minimal blast radius. They need progressive exposure, quick rollback if quality or safety metrics regress, and no big-bang cutover. Which release approach fits best?
- A regulated enterprise requires that multi-agent workloads in Azure cannot reach the public internet for tool APIs hosted on Azure App Service and Azure OpenAI. Agents still need private connectivity to those endpoints and to Key Vault. Which network design best meets the requirement?
- A banking multi-agent assistant must never recommend specific securities trades or promise guaranteed returns, even if users jailbreak the base model safety filters. Compliance wants automated proof that custom policies hold. What should you implement?
- A team maintains multi-agent definitions, tool registrations, prompt assets, and Azure infrastructure for Microsoft Foundry. They need repeatable promotions across Development, Test, Acceptance, and Production with auditability. Which approach best satisfies CI/CD and multi-environment requirements?
- In a multi-agent system, any specialist can currently invoke any registered tool, including high-risk actions such as wire transfers and customer data export. You must enforce authorization checks consistently without rewriting every agent prompt. What is the best design?
- A multi-agent system previously used one shared service principal with broad access to Key Vault, SQL, and Storage for every agent. Security mandates Zero Trust with lateral movement prevention. What identity design should you implement?
- An agent must read the signed-in user’s calendar and create events only as that user—not as a shared daemon with access to every mailbox. Which authentication approach is appropriate?
- MCP tool connectors need API keys and certificates. Developers currently paste secrets into container environment variables at build time and rebuild to rotate. What secrets-management approach should you design for multi-agent Azure deployments?
- Security wants shift-left testing of a multi-agent Foundry solution for jailbreaks, tool abuse, and data exfiltration attempts before external launch. Which capability should you incorporate into the SDLC?
- A regulated multi-agent assistant must block toxic user inputs, prevent disallowed tool calls, validate tool responses before they re-enter reasoning, and filter final answers for policy violations. What guardrail strategy matches this requirement?
- A healthcare multi-agent triage helper must never recommend specific prescription dosages, even if a model would. Built-in generic safety filters are not domain-specific enough. What should you implement?
- You are releasing a new orchestrator prompt and tool schema for a production multi-agent concierge. Risk of quality regression is moderate; you need limited real traffic exposure with fast rollback. Which release methodology fits best?
- Enterprise policy requires multi-agent solutions to move through Development, Test, Acceptance, and Production with infrastructure-as-code and a documented rollback. Which design meets this?
- Your pipeline deploys multi-agent code to Azure but only runs unit tests on pure Python helpers. Several regressions shipped because tool wiring and prompt changes were untested. What CI/CD requirement should you add?
- A multi-agent solution in Azure calls internal knowledge indexes and MCP tools hosted on private APIs. Compliance forbids agent traffic to those backends over the public internet. Which control should you specify?
- After a canary showed elevated tool-error rates, you need an immediate switch back to the last known-good multi-agent graph and model deployment with minimal DNS-level complexity. Environments already run side by side. Which approach is most appropriate?
These questions are original practice material and are NOT actual exam questions or brain-dump content. Microsoft and related marks are trademarks of their respective owners. This material is not affiliated with, endorsed by, or sponsored by Microsoft.