Practice Exam: Claude Certified Associate – Foundations

Question bank: 120 · timed mock exam: 120 min

Official exam page on the vendor's site (opens in a new tab) Official exam guide (opens in a new tab) Official prep course (opens in a new tab)

Compare: Claude Certified Architect – Foundations vs Claude Certified Associate – Foundations

  • Output Evaluation and Validation 26 questions
  • Workflow Integration and Solution Design 20 questions
  • Governance, Risk, and Responsible Use 18 questions
  • Prompting and Task Execution 16 questions
  • Product and Model Selection 14 questions
  • Configuration and Knowledge Management 14 questions
  • Troubleshooting and Optimization 12 questions
Study mode (free) Exam mode (subscribers)

Free: 25 questions in study mode, all domains, no account.

Subscription: the 120 questions of this certification, the timed exam mode, and all other certifications too.

Subscribe · €9.99 / month

Secure card payment (Stripe) · No commitment · Cancel online

How to prepare for Claude Certified Associate – Foundations with InfraPrep

The 120 questions cover 7 domains: Output Evaluation and Validation, Workflow Integration and Solution Design, Governance, Risk, and Responsible Use, Prompting and Task Execution, Product and Model Selection, Configuration and Knowledge Management, Troubleshooting and Optimization. 25 of them are free, without an account.

Study mode

Answer at your pace: each answer is corrected straight away, with an explanation and the link to the documentation it relies on.

Exam mode

A timed exam of 120 minutes, with a question map to flag and come back to questions before you submit.

Results by domain

Your score is broken down by domain, so you know what to review first.

Original questions

Every question is written from the public documentation, never copied from the exam. You can contest an answer from study mode.

10 free sample questions

Question 1 Prompting and Task Execution

A marketing coordinator types "Write a launch announcement for our new scheduling app" and gets a bland, generic draft that ignores the company's audience of hospital shift managers, its plain-spoken house style, and the 150-word limit of the newsletter slot. What is the most effective way to improve the prompt?

  1. Add the audience, purpose, house style, and length limit, as you would brief a capable new hire who lacks context
  2. Keep the prompt short and ask Claude to "be more creative and try harder" on the next attempt
  3. Send the identical prompt several more times and keep whichever draft happens to read best
  4. Move the request into an incognito chat so that earlier conversations cannot influence the result
Show the answer and the explanation

Correct answer: A

Anthropic's guidance compares Claude to a brilliant new employee who lacks context on your norms, so spelling out audience, style, and constraints is what makes the output specific.

Source: official documentation

Question 2 Output Evaluation and Validation

A procurement analyst asks Claude to summarize the obligations in a 120-page supplier agreement. Several points in the summary do not seem to match the contract. To make the next summary more accurate and easier to check against the text, what should the analyst ask for?

  1. A shorter summary, limited to five bullet points, so there is less text to check
  2. A summary written with a higher level of confidence and without hedging language
  3. A summary that avoids quoting the contract so the wording stays easy to read
  4. Word-for-word quotes of the relevant clauses first, then a summary based on those quotes
Show the answer and the explanation

Correct answer: D

For long documents, Anthropic recommends having Claude extract word-for-word quotes before performing the task, which grounds the response in the actual text.

Source: official documentation

Question 3 Product and Model Selection

A three-person agency team works on one client account for months. Every conversation needs the same brand guide, product sheets, and tone rules, and they want those chats kept together rather than mixed with unrelated work. Which Claude feature fits best?

  1. A single very long chat that everyone keeps adding new questions to
  2. Research mode, switched on at the start of every new conversation
  3. A published artifact that contains a copy of the brand guide text
  4. A project with its own knowledge base, instructions, and chat history
Show the answer and the explanation

Correct answer: D

Projects are self-contained workspaces with their own chat histories and knowledge bases, and project knowledge gives every chat in the project the same background.

Source: official documentation

Question 4 Workflow Integration and Solution Design

An operations lead is scoping a Claude workflow that summarizes incoming service tickets. When asked what success looks like, stakeholders only say that "it should work well." What should the lead do first?

  1. Start drafting prompts immediately and let the stakeholders judge the results informally
  2. Pick the most capable model available so that the quality question never comes up
  3. Define specific, measurable success criteria that the summaries can be evaluated against
  4. Ask Claude to decide for itself what a good ticket summary should contain
Show the answer and the explanation

Correct answer: C

Anthropic states that building a successful LLM application starts with clearly defining success criteria and then designing evaluations against them; good criteria are specific.

Source: official documentation

Question 5 Configuration and Knowledge Management

In a shared project, a planner agreed on final event dates with Claude in one chat. The next day, a colleague opens a new chat in the same project and Claude has no knowledge of those dates. Nothing was added to the project knowledge base. What explains this, and what is the fix?

  1. The project reached its storage limit; the owner must archive older chats first
  2. The colleague has view-only access, so Claude hides earlier decisions from them
  3. Project instructions expired overnight and must be saved again every morning
  4. Context is not shared between chats in a project; add the dates to project knowledge
Show the answer and the explanation

Correct answer: D

The help center notes that context is not shared across chats within a project unless the information is added into the project knowledge base.

Source: official documentation

Question 6 Governance, Risk, and Responsible Use

A recruiting team lists four ideas for using Claude's agentic features. Which idea conflicts with Anthropic's guidance on using agents according to the Usage Policy?

  1. Summarizing a competitor's public press releases into a monthly briefing
  2. Drafting interview schedules and sending calendar holds that the team approves
  3. Tracking a shortlisted candidate's online activity without notifying them
  4. Turning the team's interview notes into a structured comparison table
Show the answer and the explanation

Correct answer: C

The guidance prohibits using agents to monitor or track individuals' online activities, behaviors, or movements without notification or consent.

Source: official documentation

Question 7 Troubleshooting and Optimization

Claude's answer to a long request for a training handbook stops partway through a section. What does Anthropic's help center recommend when a response is too brief or only partially complete?

  1. Give Claude the previous prompt and response in your next prompt so it can continue
  2. Wait a few hours for the usage limit to reset and then send the same prompt again
  3. Turn on incognito mode so the whole context window is freed for the output
  4. Switch the conversation language to English, where outputs are always longer
Show the answer and the explanation

Correct answer: A

The article recommends retooling the prompt and giving Claude the previous prompt and response so it can pick up where it left off.

Source: official documentation

Question 8 Prompting and Task Execution

An HR analyst wants Claude to compare three regional leave policies, flag conflicts with the new global policy, and then produce a polished briefing for the leadership team. A single long prompt keeps producing a briefing that skips half of the conflicts. Which approach best applies task decomposition?

  1. Ask for the finished leadership briefing again, but in a much longer single prompt with every requirement in capitals
  2. Break the job into steps: list the conflicts first, have Claude review that list against the criteria, then draft and refine the briefing
  3. Ask Claude to produce the briefing and the conflict analysis in two different languages so it cross-checks itself
  4. Remove the three regional policies from the prompt so that Claude can focus only on writing a polished briefing
Show the answer and the explanation

Correct answer: B

Anthropic recommends breaking complex requests into substeps, and describes a draft, review, refine chain as the most common way to structure such work.

Source: official documentation

Question 9 Output Evaluation and Validation

In a regular chat with no connectors enabled, a project manager asks Claude to finalize a status update. Claude replies that it has "emailed the update to the steering committee" and includes a shared-folder link that returns an error. What is the best explanation?

  1. The email was sent but delayed, because Claude queues outgoing messages until the chat is closed
  2. The link failed because the steering committee has not yet granted Claude read access to the folder
  3. Claude hallucinated its capabilities; it has no email or file-transfer tool unless one is explicitly integrated
  4. Claude used a built-in mail service that works only for recipients inside the same organization
Show the answer and the explanation

Correct answer: C

The help center explains that Claude can hallucinate its capabilities and does not have access to email, word processors, or file transfers that are not explicitly integrated.

Source: official documentation

Question 10 Product and Model Selection

A communications team wants a default model for everyday drafting and analysis. They do not need the slowest, most capable option, nor the absolute fastest one. In Anthropic's current models overview, which model is described as "the best combination of speed and intelligence"?

  1. Claude Haiku 4.5
  2. Claude Sonnet 5
  3. Claude Opus 5.5
  4. Claude Fable 5.1
Show the answer and the explanation

Correct answer: B

The models overview describes Claude Sonnet 5 as the best combination of speed and intelligence.

Source: official documentation

Frequently asked questions

Are there free questions, and do I need an account?

Yes. 25 questions of this certification are free in study mode, across all its domains, without a subscription and without registration. The account is created when you subscribe, at checkout.

How does the subscription work, and can I cancel?

One monthly subscription covers all certifications and the timed practice exams. There is no minimum period. You cancel online, at any time, with the "Cancel your contract" function in My account; cancellation takes effect at the end of the month already paid, and no further payment is taken.

Where do the questions come from?

Every question is original, written from the vendor's public documentation. Each answer comes with an explanation and a link to the documentation page it relies on. InfraPrep is not affiliated with the vendors, and certification names belong to their owners.

In which language are the questions?

The questions are written in English, also on the French version of the site.

Is the catalog updated?

Yes, the catalog changes over time: new certifications and new questions are added.