CCAO-F Prompting and Context Engineering Practice Question
Exhibit
SYSTEM: You are a helpful assistant. USER: What is 2+2? ASSISTANT: 4 USER: And what is the capital of France? ASSISTANT: Paris USER: What was my first question?
Refer to the exhibit. Why might the model struggle to answer the final question if you are not using a stateful chat implementation?
⚠ Common exam trap
Candidates incorrectly believe the model has a persistent 'session' memory. They fail to realize that each API call is isolated and requires the full history to maintain context.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
The model's internal memory resets every time you send a new API request.
In a stateless API architecture, the model does not automatically remember previous turns. To ensure the model has access to the full conversation history, you must include the full chat thread in every API request. If you only send the most recent user message, the model loses the context of previous messages, making it impossible to answer questions that depend on historical interactions.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
The model's internal memory resets every time you send a new API request.
Why this is correct
The Anthropic API is stateless. Each request is a standalone interaction. To maintain the appearance of a conversation, you must send the entire message history (user and assistant turns) in the 'messages' parameter of the API request so the model can see the full thread and provide consistent answers.
- ✗
The model has a maximum token limit that prevents it from remembering history.
Why it's wrong here
While there is a maximum context window, it is very large (hundreds of thousands of tokens). A few short exchanges do not hit this limit. The problem is not token capacity; it is the stateless design of the API, which requires the user to manage the history explicitly.
- ✗
The system prompt is preventing the model from accessing past information for safety.
Why it's wrong here
System prompts do not restrict memory or history. The issue is purely that each API call is independent by design. Safety filters are a separate mechanism and would not interfere with the model's ability to see previous messages if they were properly included in the request body.
- ✗
The model is not optimized for question-answering tasks and lacks long-term recall.
Why it's wrong here
Claude is highly optimized for chat and question-answering. The limitation is architectural, not a performance issue of the model itself. By correctly providing the history as an array of message objects, the model will easily be able to recall information from previous turns in the chat.
About these practice questions
One of 259 original CCAO-F practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. Learn why practice questions differ from exam dumps →
JA
Written and reviewed by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
Last reviewed September 2026 · checked against the official Anthropic exam blueprint
This CCAO-F practice question is part of Courseiva's free Anthropic certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the CCAO-F exam.