AIML Rounds

QUIZ

Agents and MCP

12 questions on Agents and MCP, from 24 lessons. Pick an answer in your head, then open it.

On the roadmap for: AI Engineering.

  1. Question 1 of 12

    An agent succeeds at each step 95% of the time. Roughly how often does it complete a 20-step task with no mistakes?

    1. A. About 95%, the same as each step
    2. B. About 75%, since most steps succeed
    3. C. About 5%, since errors multiply fast
    4. D. About 36%
    View answer

    ✓ D. About 36%

    0.95²⁰ ≈ 0.36. Small per-step error rates compound over long tasks, which is why workflows or checks help.

    Lesson: What Is an Agent? The ReAct Loop

  2. Question 2 of 12

    What is the core loop of a ReAct-style agent?

    1. A. Think, call a tool, observe, and repeat until done
    2. B. Ask the user for permission at every single step
    3. C. Answer once from memory, then stop
    4. D. Retrieve documents, then fine-tune on them
    View answer

    ✓ A. Think, call a tool, observe, and repeat until done

    The model reasons, acts through tools, and uses what it observes to decide the next step.

    Lesson: What Is an Agent? The ReAct Loop

  3. Question 3 of 12

    Why does giving an agent 100 tools often make it worse?

    1. A. Each tool call is slow, so more tools mean more waiting
    2. B. Models can only ever call one tool per conversation
    3. C. Tools have to be called in the order they are listed
    4. D. Long tool lists eat context and make choosing harder
    View answer

    ✓ D. Long tool lists eat context and make choosing harder

    Fewer, well-described tools, or loading tools on demand, keeps selection accurate.

    Lesson: Tool and Function Calling

  4. Question 4 of 12

    When does asking an agent to 'check its work' help most?

    1. A. Only when the answer is a single short sentence
    2. B. Never: models can't judge their own work at all
    3. C. Always, by about the same amount for every kind of task you give it
    4. D. When there's an outside signal, like tests or a calculator
    View answer

    ✓ D. When there's an outside signal, like tests or a calculator

    Self-review with no new information often just repeats the same mistake; verification against real feedback works.

    Lesson: Planning and Reflection

  5. Question 5 of 12

    How can an agent reliably remember a user's peanut allergy 50 turns later?

    1. A. Hope the fact stays somewhere in the context window
    2. B. Save key facts to memory and bring them back when relevant
    3. C. Repeat the allergy in every reply to the user
    4. D. Switch to a bigger model with a longer context window that holds everything
    View answer

    ✓ B. Save key facts to memory and bring them back when relevant

    Context is limited and summaries lose details, so important facts are stored and retrieved deliberately.

    Lesson: Memory and Context Engineering

  6. Question 6 of 12

    What is the main idea behind agent 'skills'?

    1. A. Procedure files loaded only when needed, saving context
    2. B. Giving the agent admin rights to run any program
    3. C. Fine-tuning a separate model for each task the agent might have to do
    4. D. Hard-coding every tool call in advance
    View answer

    ✓ A. Procedure files loaded only when needed, saving context

    A short description is always visible, and the full instructions load on demand, so the agent can have many skills cheaply.

    Lesson: Agent Skills and Self-Improving Context

  7. Question 7 of 12

    How does MCP differ from plain function calling?

    1. A. MCP replaces the LLM with a fixed set of tools
    2. B. MCP only works with one vendor's family of models, so it isn't portable
    3. C. They are two different names for exactly the same thing
    4. D. Function calling asks for a tool; MCP is how tools are offered
    View answer

    ✓ D. Function calling asks for a tool; MCP is how tools are offered

    With MCP, a tool is built once as a server and works across many agents and apps.

    Lesson: MCP: The Model Context Protocol

  8. Question 8 of 12

    Your agent needs tools from your own systems and help from another company's agent. Which protocols fit?

    1. A. MCP for both the tools and the other company's agent
    2. B. Plain HTTP calls with no shared protocol at all
    3. C. No protocol: the two agents can share memory directly
    4. D. MCP for tools, and an agent-to-agent protocol like A2A
    View answer

    ✓ D. MCP for tools, and an agent-to-agent protocol like A2A

    Tools and peer agents are different connections, and each has its own protocol.

    Lesson: Agent Protocols: MCP, A2A, AG-UI and Commerce

  9. Question 9 of 12

    When do multiple agents usually beat a single agent?

    1. A. When the work splits into parts needing separate contexts
    2. B. Always: more agents always means a smarter system
    3. C. For short, simple tasks where speed matters most
    4. D. Whenever you want to lower the total token cost of a long-running task
    View answer

    ✓ A. When the work splits into parts needing separate contexts

    Multi-agent setups cost many more tokens and add coordination problems, so they pay off on wide, parallel work.

    Lesson: Multi-Agent Systems

  10. Question 10 of 12

    A long agent run crashes halfway after charging a customer. How do you avoid charging twice on restart?

    1. A. Retry the whole run from the very beginning
    2. B. Turn off payments for long-running agents
    3. C. Durable execution with checkpoints and idempotency keys
    4. D. Ask the model whether it already charged the customer
    View answer

    ✓ C. Durable execution with checkpoints and idempotency keys

    Recorded steps are skipped on resume, and an idempotency key makes a repeated charge a no-op.

    Lesson: Agent Infrastructure: Durable Execution, Retries and Budgets

  11. Question 11 of 12

    What is the 'lethal trifecta' for an AI agent?

    1. A. Being slow, expensive and wrong at the same time on the same request
    2. B. Letting three agents talk to each other
    3. C. Giving the agent three tools at once
    4. D. Private data, untrusted content and a way to send data out
    View answer

    ✓ D. Private data, untrusted content and a way to send data out

    With all three, a hidden instruction in an email or web page can make the agent leak data. Remove at least one leg.

    Lesson: Agent Safety: Permissions, Sandboxing and the Lethal Trifecta

  12. Question 12 of 12

    Why do human approvals often fail to catch an agent's mistakes?

    1. A. Humans are slower than agents, so they fall behind
    2. B. Too many requests, too little context: people rubber-stamp
    3. C. Approvals are usually done automatically in practice, so nobody checks them
    4. D. Agents hide their real requests from the reviewers
    View answer

    ✓ B. Too many requests, too little context: people rubber-stamp

    Ask for approval only on risky actions, and show what will change and why, so each review is meaningful.

    Lesson: Human-in-the-Loop Design for Agents