Claude Certified Associate Practice Test

Answer: C
Configuration quality depends on maintained knowledge. Project knowledge is read by every chat in the project, so the fix is at the source: option C. Option D instructs Claude to treat superseded policy as current, which turns a stale answer into a deliberate one. Option A keeps both versions and hopes retrieval prefers the newer; RAG ranks passages by relevance to the question, not by which document is current, so either version can be returned. Option B clears memory, which is not where the handbook lives; the outdated file would still be retrieved on the next question.

Answer: B
Effective system-level instructions, a Domain 5 objective, set durable process rules: which definitions govern, how to handle missing data, how to separate fact from opinion, and what structure to follow; that is option B. Option A optimizes for impression and lets the metric set and layout drift, so weeks cannot be compared. Option C forbids the clarification that missing or ambiguous data requires. Option D instructs Claude to fabricate figures and hide that they are estimates, which is the hallucination risk an instruction should prevent.

Answer: B
Domain 5 covers managing connectors and knowledge sources: access, authority, and retrieval quality all need checking before a workflow depends on them; that is option B. Option A confuses reachability with approval; inheriting the owner's permissions means the connector can see drafts and superseded copies too. Option C widens access and removes the approval step at the moment the folder structure is known to be ambiguous. Option D uses the private-project setting to hide provenance from the people who must review outputs, the opposite of good governance.

Answer: D
Domain 5 asks associates to inform, maintain, and update configurations. Instructions, knowledge, and permissions drift out of date as policies, sources, users, and goals change, so option D combines a calendar with event triggers. Option C waits for harm before looking. Option B confuses persistence with correctness; saved instructions stay exactly as written while the business moves on. Option A ties review to vendor events only; a new model or interface is one trigger, but a changed policy or source document is the more common reason a project starts giving wrong answers.

Answer: B
Auto memory is built as Claude works and saves learnings across that developer's sessions, but it is not the same as the shared, versioned CLAUDE.md file every teammate's session reads at the start; carrying a discovery to the whole team still means writing it into CLAUDE.md, which is option B. Option D describes an automatic sync from memory to CLAUDE.md that does not exist; the file is a maintained document, not a generated log. Option A invents a Skills behavior that has no basis in how memory or Skills work. Option C is simply false about what auto memory can hold; nothing restricts it to pricing or context-limit facts.

Answer: C
Enterprise plans provide admin controls, usage analytics, and observability, including an Analytics API and OpenTelemetry, over Cowork sessions, which is the built-in way to get org-wide visibility: option C. Option A substitutes manual, unreliable self-reporting for the tooling built for this purpose and does not scale across hundreds of employees. Option B removes connector access altogether, which stops the workflows the officer is trying to observe rather than monitoring them. Option D assumes no configuration is available or needed, when Enterprise specifically adds the oversight tooling this scenario calls for.

Answer: D
CLAUDE.md is a markdown file at the project root that Claude Code reads at the start of every session, and it is meant for exactly this: architecture decisions and preferred libraries the whole team should follow, which is option D. Option C leaves the fix to auto memory, which is built during a session as Claude works and is not guaranteed to capture and generalize a specific past decision the way a written standard does. Option A only reaches whichever chat happens to raise the topic again and leaves every other session, and every other engineer, without the standard. Option B depends on a new hire finding and correctly interpreting an old chat thread instead of Claude Code itself being configured to know the rule.

Answer: A
Domain 6 requires following organizational AI policies and escalating uncertainty. Option A stops the non-compliant use and routes the question to whoever can approve or amend the policy. Option D substitutes personal judgment about usefulness for the governance process. Option B treats the model as the policy authority; Claude can summarize a policy, but it cannot grant an exception. Option C moves company work to a consumer account where organizational retention controls, admin oversight, and commercial terms no longer apply, which increases the risk the policy exists to manage.

Answer: C
Journalism is one of the Usage Policy's High-Risk Use Cases, requiring a qualified professional's review before content is disseminated and disclosure to the audience that AI was involved, which option C satisfies. Option A checks the draft against style, not the factual review the policy requires. Option D has the model verify its own draft, which is not the independent qualified review the policy calls for — a model checking itself is still model output. Option B discloses only reactively, after publication and only if a complaint arrives, rather than to readers before or at publication as the policy requires.

Answer: A
Domain 6 applies data-sensitivity and privacy controls to the data itself. Option A removes or masks the regulated identifiers before they leave the controlled system, which is what the policy requires. Option B cites a true training exclusion, but the policy restricts sharing the data at all, and no-training is not the same as permitted disclosure. Option C misuses incognito mode; it keeps a chat out of history and memory, it does not make an upload of regulated data compliant. Option D obscures labels while the names and numbers remain in the file.

Answer: D
Employment decisions are a high-risk use under Anthropic's Usage Policy: a qualified person must review the output before it is acted on, and affected people must be told AI was involved. Historical hiring data can encode past bias, so option D examines data, criteria, and legal exposure and keeps a human in the loop. Option C treats biased outcomes as ground truth. Option A hides the method, which defeats the disclosure and review the policy requires. Option B makes the model the sole decision-maker, exactly what the human-in-the-loop requirement prohibits.

Answer: D
Healthcare content is a high-risk use case under Anthropic's Usage Policy, which requires a qualified professional to review such material before it is disseminated. Option D grounds the draft in approved sources and adds that clinical review. Option A treats fluent wording and the presence of citations as accuracy; neither is. Option C shifts verification to the public after the risk has already been created. Option B removes the trail a reviewer would use to check claims, making the review harder rather than the handout safer.

Answer: B
Academic testing is listed among the Usage Policy's High-Risk Use Cases, so a qualified professional — here, an educator — must review the AI-generated score and placement before it is finalized, and students must be told AI is being used, at the start of the process rather than buried elsewhere, which is option B. Option A only reviews after a parent appeals, which is review after finalization, not before. Option C skips review for a subset of students based on their own score, which is not a review at all for that group. Option D places the disclosure in a handbook distributed months before the exam is even scored, not tied to the actual use of AI on that student's exam, which does not satisfy an upfront, connected disclosure.

Answer: A
Anthropic's prompting guidance recommends giving Claude a role in the system prompt, even a single sentence, to focus tone and behavior for a specific use case; that is option A. Different requesters produce different drafts because nothing anchors the voice, and a role prompt fixes that anchor regardless of who asks. Option D adds reasoning depth, which affects thoroughness, not voice consistency. Option C raises effort, a cost and latency lever, not a tone control. Option B pushes the inconsistency back onto the team instead of fixing the prompt, and still leaves no shared voice for Claude's own drafts.

Answer: B
Anthropic's prompting guidance opens with 'Be clear and direct' and 'Add context to improve performance': a generic draft is the symptom of a prompt that never said who the email is for, what it must achieve, or what the deliverable looks like. Option B supplies exactly those missing inputs. Option A adds emphasis, not information, and the guidance for current models is to dial back capitalized MUST/CRITICAL language. Option C asks for reasoning depth and volume, neither of which fills a context gap. Option D removes the little direction the prompt had, so the draft drifts further from the campaign.

Answer: D
Task decomposition is a Domain 1 objective: a request that asks for theme extraction, cross-role comparison, and an executive brief in one turn gives Claude no checkpoint, so parts get dropped. Option D sets a deliverable per stage that the associate reviews before the next stage starts. Option C repeats the same overloaded request and changes nothing about why it fails. Option A discards the evidence the analysis exists to examine, so any themes would be invented. Option B removes the source material rather than the overload; a brief written without the transcripts cannot be checked against them.

Answer: B
Iterating a prompt means giving diagnostic feedback: name what is missing and the requirement it must meet, then ask for the revision. Option B does that and hands Claude the policy language to satisfy. Option D substitutes self-assessment for review; a model's confidence rating is not evidence that the two points were covered. Option A throws away the working structure and the feedback, and a fresh context has no reason to produce the omitted points. Option C changes register, not substance; the two mandatory points stay absent in a friendlier draft.

Answer: D
Domain 1 asks candidates to adapt the prompting strategy to the task type. Brainstorming calls for divergent output: several options, each tied to the need it serves, with the choice deferred to the human, which is option D. Option B converges immediately and hides the alternatives the associate wanted to compare. Option C is drafting work for a decision that has not been made, so the memo would rest on a theme nobody chose. Option A strips the reasoning that makes options comparable; bare scores say nothing about why a theme fits the audience.

Answer: B
Examples are one of the most reliable ways to steer format, tone, and structure; Anthropic's guidance recommends three to five relevant, diverse examples wrapped in example tags, which is option B. Option A describes the target in the abstract, and abstract description is a weaker steering signal than demonstrated examples. Option D states one adjective, professional, and leaves structure undefined, so postings will still vary by recruiter. Option C asks Claude to imagine a standard without ever showing one, which reintroduces the same inconsistency the agency is trying to remove.

Answer: C
Wrapping distinct content types, instructions, context, examples, and input data, in their own clearly named tags keeps Claude from conflating them, which is option C. Option D removes the sample clause instead of separating it, losing the example the analyst wanted to demonstrate. Option A repeats the instructions but leaves the contract and the sample clause running together in the same untagged block, so the conflation persists. Option B adds a rereading instruction without changing how the content is structured, which does not address why the two texts got mixed.

Answer: D
Manual step-by-step reasoning helps most on tasks with several intermediate steps that are easy to skip, such as multi-line variance math, and asking Claude to check its own answer against the stated figures catches errors before they reach the final table; that is option D. Option C adds a role, which shapes tone and framing, not arithmetic accuracy. Option B shortens the answer, which removes the room needed to show and check each calculation. Option A reruns the same unstructured request and hopes for a better outcome by chance, rather than changing what would cause the subtotals to be checked.

Answer: A
Anthropic's guidance is to tell Claude what to do rather than what to avoid; stating the desired positive form, flowing prose paragraphs, is a clearer format instruction than a list of prohibitions, which is option A. Option B lists three things not to do without saying what the output should look like instead, which is the weaker framing the guidance warns against. Option D changes reasoning depth, not formatting, and list structure can persist regardless of the thinking setting. Option C accepts the wrong format first and asks for a manual conversion afterward, adding a step instead of preventing the mismatch.

Answer: B
Anthropic's current guidance is that heavy capitalized emphasis, such as MUST and CRITICAL, can cause behaviors to overtrigger on current models, and normal phrasing that states the actual condition works better; that is option B. Option A adds more of the emphasis that is already causing the overtriggering, which would make the problem worse, not better. Option D removes the escalation criteria entirely, so Claude has no basis for flagging real delays either. Option C asks for an internal state change Claude cannot perform through a prompt instruction; the fix is the wording of the criteria, not a request to disregard training.

Answer: C
Anthropic's long-context guidance places long documents and inputs near the top of the prompt, above the query and instructions, because queries placed at the end can measurably improve response quality on multi-document and long-document tasks; that is option C. Option D repeats the question rather than repositioning the transcript, so the document still follows the query. Option B adds a summary but keeps the transcript after the question, the ordering the guidance advises against. Option A shortens the question without moving the transcript, so the structural problem the paralegal is seeing remains.

Answer: D
For long-document tasks, asking Claude to first quote the relevant passages, for example into a dedicated quotes section, before doing the analysis grounds the response in the source text and reduces drift; that is option D. Option A reverses the order, answering from memory first and checking only when uncertainty is already suspected, which is exactly the pattern the quoting technique prevents. Option C shortens the answer without adding any grounding step, so a short wrong answer is just as likely as a long one. Option B summarizes the whole ordinance broadly, which does not connect a specific detailed question to the exact clause that answers it.

Answer: A
Prefilled assistant responses are no longer supported starting with Claude 4.6-generation models, and a request with a prefilled last-turn message returns a 400 error on those models; the documented migration is to use Structured Outputs to constrain JSON output instead, which is option A. Option C still relies on a prefilled final-turn trick, just relocated, and does not address why the newer model rejects it. Option D avoids the error by giving up the newer model's capabilities rather than adopting the supported replacement. Option B retries a request that will keep failing for a structural reason, not a transient one.

Answer: B
Since prefilling the last turn is no longer supported on current-generation models, the documented replacement for skipping a preamble is a direct system-prompt instruction, such as asking Claude to respond directly without preamble, which is option B. Option D is another form of the unsupported prefill approach, just phrased as an unfinished sentence, and would fail the same way. Option A asks Claude to generate the unwanted preamble and then remove it itself, an extra step that does not stop the preamble from being produced. Option C changes reasoning depth and overall length, not the presence of an opening pleasantry.

Answer: C
A single request that asks for extraction, comparison, and flagging across twenty-five contracts overloads one turn with no checkpoint, so parts get dropped partway through; breaking the request into staged sub-tasks with a defined output and a review point at each stage produces a more complete result, which is option C. Option D shortens only the final report, not the multi-part task that is failing to finish. Option B trades completeness for speed on the very documents the review depends on. Option A repeats the identical overloaded request twice and hopes the gaps do not line up, rather than changing the structure that is causing them.

Answer: D
Anthropic's clarity guidance is to give sequential, numbered steps in the instructions whenever order matters, since Claude otherwise has to infer sequence rather than follow a stated one; that is option D. Option A asks Claude to infer order from general knowledge instead of the coordinator supplying the actual required sequence, the gap causing the misplacement. Option C shortens the procedure but does not supply the ordering information that is missing. Option B removes the safety check that keeps getting misplaced rather than fixing its position, which drops a required step from the procedure entirely.

Answer: A
Anthropic's stated golden rule is that if a colleague with minimal context would be confused by an instruction, Claude will likely be confused too; 'make this better' gives no specifics, so option A supplies the missing problems, constraints, and audience the vague prompt omitted. Option C literally invokes the golden rule's phrasing but still never states what to fix or preserve, so the confusion remains. Option B repeats the identical vague instruction and gives Claude no new information to act on. Option D removes the only instruction that existed, leaving Claude to guess the goal from the draft alone.

Answer: B
Examples should be diverse enough to cover the range of real cases without teaching an unintended pattern; a set that only shows angry callers teaches Claude that every reply opens with an apology, so option B adds calmer, neutral examples to correct the imbalance. Option D shrinks the set but keeps only the angry-caller pattern, so the same unintended generalization persists. Option C keeps the skewed examples and adds a prohibition that fights what the examples are actually demonstrating. Option A removes the examples, which are one of the most reliable tone-steering tools, instead of fixing what they demonstrate.

Answer: C
Asking Claude to check its own answer against stated criteria before finishing reliably catches errors, especially on coding and math tasks, so option C directs that verification at the exact edge case the patch is still failing. Option A discards the specific failure case for a fresh approach, with no guarantee it addresses that case either. Option D asks for brevity, which does not confirm whether the edge case now passes. Option B produces an explanation for a different audience, not a check of whether the patch actually handles the reported case.

Answer: D
Anthropic's guidance on adding context is that explaining why an instruction matters, such as noting that text-to-speech will read the notes aloud, helps Claude generalize better than a bare rule alone, which is option D. Option B mislabels the technique; stating a reason is the 'add context' practice, distinct from assigning Claude a role in the system prompt. Option A is not how the reason functions; a longer prompt does not by itself trigger more thinking, and thinking depth is not what fixes formatting choices like ellipses. Option C describes a numbered-steps technique, which addresses ordering, not the rationale behind a formatting constraint.

Answer: A
The first draft went wrong because the prompt never named which section mattered, so a scope-adding revision, naming the finished-goods forecast specifically, is the direct fix, which is option A. Option D changes the model but keeps the same underspecified request, so the new model has no more information about which section to summarize than the last one did. Option C expands the request to the whole report instead of narrowing it to the needed section, reintroducing the ambiguity in a different form. Option B repeats the identical vague request and hopes for a different, uncontrolled outcome rather than fixing what caused the wrong section to be picked.

Answer: B
Audience-specific criteria matter when drafts are written for different readers; here both paragraphs target the same program officer, so one shared rubric on evidence, clarity, and call-to-action is the correct comparison, which is option B. Option D applies the audience-split method to a case where there is only one audience, which would produce two criteria sets for a reader who needs just one. Option A rewards length rather than the qualities that make a case-for-support persuasive. Option C has the model that wrote both paragraphs also pick the winner, rather than scoring against defined criteria.

Answer: B
Language models can produce specific-looking details, including subsection numbers, that are wrong; that is the hallucination pattern named in Domain 2. Option B checks both the citation and the summary against the authoritative text before a compliance audience relies on it. Option A treats a tool setting as verification; web search can surface a source, but it does not confirm that this summary reads it correctly. Option C relies on self-reported confidence, which is not an accuracy signal. Option D improves presentation while leaving correctness untested.

Answer: A
Domain 2 separates accuracy from completeness. Nothing in the update is false, so the defect is that information the reader needs for the decision, the budget risk and the deadline, is missing; that is option A. Option C would require fabricated content, and none is present. Option D would require a false statement, and the update is factually accurate. Option B concerns detail level and vocabulary for a particular reader; the update's problem is omitted facts, which would be missing for any audience.

Answer: C
Domain 2 includes editing, adapting, and comparing outputs for the intended audience. Engineers need mechanism and terminology; executives need the decision, its risk, and the next action, so the yardsticks differ, which is option C. Option A rewards length, and the executive draft should usually be the shorter one. Option D outsources judgment to the model that wrote the drafts and scores prose, not fitness for a reader. Option B guarantees a wrong result for one audience, because a single rubric cannot value brevity and depth at the same time.

Answer: D
Validation tests provenance, freshness, and disagreement. Option D does all three: each material claim is traced to where it came from, its date is checked, and conflicts are surfaced rather than smoothed over. Option B tests format, not truth; a tidy table can be uniformly wrong. Option C confuses repetition with confirmation; vendor marketing is copied across sites, and two pages restating one press release are still one source. Option A removes the very trail a reviewer needs, so nothing in the table could be checked afterward.

Answer: B
A citation is not proof that the underlying claim is accurate; the subsection and what it actually authorizes must be checked against the regulation itself before the analyst relies on it. Option A treats database access as verification, which it is not — access does not guarantee the passage was read correctly. Option C substitutes the model re-asserting its own answer for an independent check against the source text. Option D adds a disclosure, which is good practice for AI-assisted advice, but disclosure does not substitute for confirming the cited subsection actually says what the summary claims.

Answer: C
A quote that sounds plausible is exactly the hallucination pattern this domain warns about: specific-looking, confidently stated content that may not be grounded in any real source. Option C is the only step that checks the quote against an actual record. Option A judges plausibility by style, not by evidence. Option B asks the model that produced the quote to grade its own accuracy, which is not independent verification. Option D changes the framing so the claim reads as unconfirmed, but a wrong quote attributed to a named person is still false regardless of the hedge word used.

Answer: D
A number that names its source is not the same as a number confirmed against that source; the memo could misread a line item or apply the wrong formula. Option D is the only action that independently reproduces the figure from the underlying financial statement. Option A treats a source citation as proof rather than a claim to be checked. Option B asks the same model to re-verify its own work, which does not catch an error in how it read the P&L in the first place. Option C avoids verifying the exact figure by blurring it, but a board deck built on an unverified number carries the same risk whether it is precise or rounded.

Answer: A
Tone and empathy are subjective quality dimensions, and the evaluation guidance names model-graded rubrics on a Likert scale as the fit for exactly this kind of criterion; that is option A. Option B is for categorical tasks with one correct answer, and a reply's wording legitimately varies while still being empathetic. Option C measures text overlap with a reference document, which says nothing about tone. Option D measures how similar two transcripts are to each other, not whether either one is empathetic or on-brand.

Answer: B
Evaluation guidance is explicit that measurable success criteria and a diverse test set, including edge cases, come before a workflow ships, not after; that is option B. Option A checks one case and calls it representative, which a report type never seen in that one sample would fail silently. Option C skips evaluation entirely based on first impression, the failure mode the test-set practice exists to prevent. Option D has the model that produced the summaries also judge them, with no independent cases to catch a systematic blind spot.

Answer: C
Sorting into one of a fixed set of categories is a categorical, single-correct-answer task, and the grading approach named for exactly that is exact match; that is option C. Option A applies a subjective quality scale to a task that has one right label, not a spectrum of quality. Option B measures word overlap, which is irrelevant to whether the assigned category is the correct one. Option D checks similarity between unrelated tickets over time, which says nothing about whether any single ticket was classified correctly.

Answer: D
Summaries are free-text and will never match a reference word for word, so the named approach for this case is automated text-overlap scoring such as ROUGE-L; that is option D. Option A demands identical text, which no acceptable paraphrase would produce, so it would fail good summaries along with bad ones. Option B only checks that output exists, not whether it resembles the reference in content. Option C grades a trait, creativity, that is not the property being tested; the marketer wants closeness to a reference, not inventiveness.

Answer: A
A specific-looking citation that cannot be located is the hallucination pattern this domain names directly: invented, confident-sounding detail with no real source behind it. Option A keeps searching the authoritative database and, failing that, marks the citation unconfirmed rather than treating absence of proof as proof. Option B invents a justification for a citation that has not been found anywhere. Option C asks the source of the possible error to also be the judge of it. Option D keeps an unsupported legal point in the memo instead of resolving whether it has any real authority behind it.

Answer: B
The requirement is a single yes/no compliance check, and the grading approach named for exactly that is a binary pass/fail rubric; that is option B. Option A scores a quality spectrum, tone, for a task that only needs a presence check. Option C measures general text overlap with an old template, which would flag emails as different even when the required statement is present and correctly worded. Option D would fail nearly every email, since acknowledgment emails legitimately vary in wording while still containing the required statement.

Answer: C
Comparing drafts against a defined rubric, rather than defaulting to length, is how a comparison stays evidence-based; that is option C. Option D rewards the longest draft on the assumption that more detail is better, when a denial letter also has to be clear and legally sound, qualities length alone does not guarantee. Option B has the model that generated the drafts also judge them, which is not independent evaluation. Option A combines every detail from three drafts into one letter, which is likely to produce a longer, less clear document rather than the strongest one.

Answer: D
A confident tone is not evidence of accuracy, and a model that is not prompted to flag uncertainty may state a guess as if it were fact; that is the reasoning behind letting Claude say it does not know rather than guess. Option D both invites that signal and still checks the date independently, which a confidence flag alone cannot substitute for. Option A treats the absence of hedging as proof. Option B softens the wording rather than resolving whether the year is right, so a wrong date can still reach readers, just phrased more cautiously. Option C compares two unverified guesses and picks by tone rather than checking either against a record.

Answer: A
For long-document tasks, asking Claude to quote the relevant passages first grounds the response in the actual text and reduces drift from the source; that is option A. Option C reverses the recommended order for long documents, where the source material should come before the query, not after. Option B adds a layer of paraphrase between the answer and the original wording, so an error in the summary would carry into every later answer. Option D gives more space for explanation but does not connect any part of that explanation back to specific text in the agreement.

Answer: B
A list of sources shows what Claude consulted, not that every claim in the report correctly reflects what those sources say; that distinction is option B. Option A treats the presence of a source list as verification, when the report could still misstate or overstate what a listed source actually said. Option C throws away exactly the trail a reviewer would use to spot-check the claims. Option D substitutes a count for a check; ten weak or repetitive sources are not more reliable than three that were actually read and matched to the claims.

Answer: C
Consistency across multiple outputs, rather than a match to one fixed answer, is what embedding cosine similarity is named for in the evaluation guidance; that is option C. Option A demands identical wording to a single reference, which penalizes any acceptable paraphrase even when tone and claims are consistent. Option B checks fidelity to the spec sheet, a different question from whether the five variants are consistent with each other. Option D checks for one field's presence, not whether tone and claims line up across the variants.

Answer: D
Structured Outputs constrains the response to conform to a given schema, which is the current, reliable mechanism for machine-parseable output; that is option D, still paired with validating the result. Option A relies on the older prefill trick, which is no longer supported on current-generation models and returns an error rather than forcing the format. Option B substitutes the model's self-report for schema validation. Option C only catches errors a reviewer happens to spot and leaves the generation method itself unconstrained, so the same mistakes recur.

Answer: A
Nothing in the summary is false and no required section is missing, so the defect is that the register and vocabulary do not match the intended reader, which is exactly what an audience-fit finding names; that is option A. Option B would require a false statement, and the summary is factually correct. Option C would require a missing required element, and every section is present. Option D would require fabricated content not grounded in any source, and the terminology, while inaccessible, is not invented.

Answer: C
A precise-looking figure is not evidence it is correct, so the analyst needs to check it against the system of record before a reorder decision depends on it; that is option C. Option A mistakes precision for verification; a wrong number can be stated just as exactly as a right one. Option B asks the same source to re-confirm its own figure rather than checking an independent record. Option D avoids the exact number without resolving whether either the precise or the rounded figure is actually correct.

Answer: D
The verification effort a draft needs scales with what in it is novel or consequential, not with how the language reads; that is option D, which targets exactly the citations, statutes, and case-specific facts that could be wrong. Option A judges by writing style, which has no relationship to factual correctness. Option B assumes reused language means the new, case-specific content added around it is also safe, which does not follow. Option C checks two sections by position in the document rather than checking the sections that actually contain unverified, case-specific claims.

Answer: A
One statement in the report is factually false against the actual dashboard figure, while every section is present and the tone fits the reader, so the precise finding is an accuracy problem, option A. Option B would require a missing required element, and nothing is missing here. Option C would require a tone or detail-level mismatch for the reader, and neither is present. Option D fits fabricated content invented with no basis; a wrong number drawn from misreading a real dashboard is a factual error grounded in a real source, which is more precisely an accuracy defect than an invented one.

Answer: B
Evaluation guidance calls for defining measurable success criteria and testing against them before something ships, rather than approving on a read-through; that is option B, applied to both versions before either is used. Option A substitutes one person's impression for a defined check. Option C uses a vague, subjective feeling and a prior year's tone as the bar, neither of which is a measurable criterion. Option D treats internal distribution as low-risk enough to skip evaluation, but a board memo with inaccurate or incomplete figures carries its own real consequences.

Answer: C
Materials, dimensions, and price are precise, categorical fields, which is exactly what exact-match checking suits, and sampling first with expansion on failure keeps the check scalable; that is option C. Option A and option B both use a surface trait, photo count or word count, that has no logical connection to whether the stated facts match the spec sheet. Option D has the model that produced the descriptions also judge whether they drifted from the source, which is not an independent check.

Answer: C
The verification an output needs scales with the stakes and novelty of what it will be used for; a memo feeding a signed opinion needs its interpretation checked against the actual standard and firm guidance, which is option C. Option D treats a section citation as if citing it were the same as it being applied correctly. Option B treats internal logical consistency as proof of correctness, when a reasoned argument can still start from a misreading of the standard. Option A makes verification contingent on someone else happening to notice a problem, rather than being a defined step before reliance.

Answer: A
Comparing two outputs meaningfully starts with defined criteria applied to both, which is option A: accuracy against the product documentation, whether every step is covered, and whether it reads clearly. Option B uses length as a proxy for quality, which can favor an article that is shorter because it skipped a step. Option C collects opinions with no shared standard, so agents could be reacting to different things. Option D uses a search-ranking signal that reflects indexing and past traffic, not whether the article's content is correct or complete.

Answer: B
A specific number attributed to an earnings call needs to be checked against the actual transcript or release before an internal briefing relies on it; that is option B. Option A treats professional-sounding prose and a named source as proof, neither of which confirms the figure is right. Option C changes the attribution to something vaguer without resolving whether the number itself is correct. Option D substitutes the model's self-reported confidence for an independent check, and a stated high confidence is not evidence the figure is accurate.

Answer: C
Naming a credible source is not the same as the figure correctly reflecting what that source says, so the number needs to be checked against the actual estimate before a client briefing depends on it; that is option C. Option A accepts the attribution as sufficient without reading the underlying estimate. Option B avoids repeating a specific number without resolving whether it, or any other figure, is correct. Option D asks the model to repeat its earlier claim, which confirms consistency with itself, not accuracy against the budget office's actual publication.

Answer: B
Claude Code is the agentic coding tool that reads a codebase, edits files, runs commands, and integrates with dev tools, available across the terminal, IDE extensions, a desktop app, and the web, which is exactly this bug-fix workflow: option B. Option A's Project knowledge base lets chats reference uploaded files, but it does not edit the repository or run the test suite; the developer would still do that work by hand. Option C describes Cowork accurately as a working session in a user's files and connected apps, but Anthropic distinguishes Cowork's general office work from Claude Code's developer-focused agentic coding inside a codebase, which is what running a test suite requires. Option D's artifacts support in-place editing and version history for content in the chat window, but they do not execute commands or run a test suite against a repository.

Answer: C
Anthropic describes Claude Cowork as 'Claude Code for the rest of your work,' built to work directly in a user's files, folders, and connected apps such as Google Drive, Gmail, and DocuSign rather than a chat back-and-forth, which matches this multi-app task: option C. Option A names a real connection method, MCP, but Claude Code is positioned as a developer-focused agentic coding tool for a codebase, not the office-file working session this manager needs. Option B loses the direct file-and-app action Cowork is built for, and pasting dozens of contracts one at a time does not scale to reorganizing a folder or routing signatures. Option D's Project sharing permissions govern who can view or edit a project's knowledge; the described capability to act on files and route them through DocuSign as a working session belongs to Cowork, not a Project's connector.

Answer: D
Tool use lets the API call functions the application defines: the backend sends a tools array, Claude responds with a tool_use block naming the function and arguments, the backend executes it, and returns a tool_result, all without a person opening a Claude app; that is option D. Option A's Project knowledge is read within chat conversations by people using the app, not invoked programmatically from a separate backend's automatic reply loop. Option B's Cowork is a working session inside Anthropic's own desktop, web, or mobile app acting on a user's files and connected apps, not a headless component embedded in another system's request-response cycle. Option C's Routines run recurring Claude Code jobs such as a scheduled PR review inside a codebase; they are not built for a per-ticket backend integration returning replies to a support system.

Answer: A
Prompt caching is built for exactly this pattern: a stable long system prompt, tool set, or reference document reused across many requests with only the final turn changing, cached at a breakpoint and read back at roughly a tenth of the base input price; that is option A. Option B's Batch API discount is about tolerating asynchronous processing, not about the prompt repeating; it does not specifically target the reprocessing cost of an unchanged prefix, and this workflow returns replies per request rather than overnight. Option C changes model size, but every model still reprocesses the full repeated prefix at full price without a cache breakpoint, so a smaller model does not remove the repeated-processing cost. Option D is not model-specific: any of the current models supports caching, and the 1-hour TTL is simply the higher-priced alternative to the 5-minute default, not a Fable 5.1 exclusive.

Answer: A
Domain 3 aligns model selection with cost, speed, and quality. Anthropic's model guidance recommends starting efficiency-first for high-volume, straightforward tasks: begin with Claude Haiku 4.5, test on a sample, and upgrade only if a capability gap appears, which is option A. Option C pays the highest per-token price and the slowest latency in the lineup for drafts that need no deep reasoning. Option D chooses by tier label rather than by measured fit. Option B picks a plausible model but skips the evaluation that proves it meets the threshold.

Answer: C
Projects hold two durable things: project instructions and a project knowledge base that every chat in the project reads, which is exactly what recurring research with a stable brief needs; hence option C. Option A leans on memory, which saves topics Claude notices rather than a curated document set, and each project has its own memory space anyway. Option B rebuilds the reference set from the open web on every run, so the sources drift and are never maintained. Option D publishes confidential documents into a shareable artifact, a data-handling failure, not a workspace.

Answer: B
Domain 3 covers context limits and memory: knowing when to restart, summarize, or persist. Stale assumptions in a long thread keep steering answers, so option B carries forward only what is still true and states the task fresh. Option D mistakes capacity for relevance; a larger context window still contains the outdated turns and gives them weight. Option A adds more history, including more stale material, to the problem. Option C removes the requirements that define the task while leaving the outdated assumptions in place, which inverts the fix.

Answer: D
Domain 2 includes selecting output formats: artifacts, inline text, or structured data. Claude creates an artifact for content that is substantial, self-contained, and likely to be edited or reused; a quick inline answer suits a one-time reply. So the deciding factor is option D. Option B is appearance, not workflow. Option C forces one format onto tasks with different review and reuse needs. Option A is the wrong goal: a stakeholder deliverable should be reviewed regardless of format, and the format choice does not change that duty.

Answer: C
Anthropic's own model guidance says to start with Claude Opus 5 for most workloads and move to Claude Fable 5.1 for demanding reasoning and long-horizon agentic work, or when evals on Opus 5 at higher effort still fall short, which is exactly this team's evidence. Option A moves to Sonnet 5, the model built for everyday speed and cost efficiency, a step down in reasoning depth from the model that already fell short. Option B moves to Haiku 4.5, the fastest and cheapest model with a smaller 200K context window and no adaptive-effort control, even less suited to a long, interdependent refactor. Option D removes thinking rather than adding capability; Opus 5 only allows disabling thinking at effort high or lower, and cutting the deliberation the eval showed was needed does not recover missed steps.

Answer: D
Anthropic's model-selection principle is to start with the fastest, cheapest model that meets a tested quality bar and upgrade only if a capability gap appears, rather than defaulting to the top-tier model by name alone. The sample review found exactly that gap, so option D tests the next tier up. Option A keeps the model that already failed the eval; more examples do not add the reasoning depth Haiku 4.5's manual, non-adaptive thinking lacks for ambiguous escalation cues. Option B skips the incremental step and pays Fable 5.1's $10/$50 per-MTok rate, the highest in the lineup, before confirming a mid-tier model cannot close the gap. Option C's batch discount lowers cost per token; it does nothing to fix a demonstrated accuracy problem in what the model decides to flag.

Answer: D
Fable 5.1, Opus 5, and Sonnet 5 all share a 1M-token context window, so the extraction fits any of them, and the task is a straightforward pass rather than demanding reasoning, so Sonnet 5's lower price is the fitting choice: option D. Option B fails outright: Haiku 4.5's context window is 200K tokens, well short of 700,000, so it cannot hold the archive at all no matter how cheap or fast it is. Option C treats Opus 5's status as a good general starting point as a requirement for large documents, but Sonnet 5's window already covers the archive at a lower per-token price. Option A applies the guidance for demanding, long-horizon agentic work to a single-pass extraction that needs neither Fable 5.1's reasoning depth nor its $10/$50 per-MTok price.

Answer: B
The Batch API prices at about half the base per-token rate specifically for workloads that can tolerate asynchronous processing, and an overnight batch with no one waiting on an individual reply is that case exactly: option B. Option A pays the standard real-time rate on the most expensive model in the lineup for a task with no latency requirement, the precise situation batch pricing exists to discount. Option C's caching discount applies to a repeated shared prefix read from cache; ten thousand distinct, independent line items have no long common prefix to cache, and its discount margin is not interchangeable with the Batch API's. Option D's usage tier governs requests-per-minute, input-tokens-per-minute, and output-tokens-per-minute limits, not the per-token price, so raising it would not lower the bill.

Answer: C
On current models including Sonnet 5, thinking is adaptive: the model decides whether and how much to think based on an effort setting and the query's complexity, with no fixed budget for the engineer to manage; that is option C. Option A's manual budget_tokens cap is the older, deprecated mechanism, not the current one on Sonnet 5. Option B's prefilled assistant response is no longer supported starting with Claude 4.6-generation models and returns an error rather than starting a reasoning block. Option D's manual chain-of-thought prompting is described as a fallback for models or configurations where thinking is off, not the most reliable technique on a model whose adaptive thinking is already available.

Answer: D
Claude Haiku 4.5's thinking is extended and manual, not adaptive, and it does not support an effort parameter at all, so the team's plan will not work on this model as written: option D. Option A invents a silent fallback to an adaptive default that Haiku 4.5 does not have, since it has no adaptive thinking mode to fall back to. Option B is factually wrong: Haiku 4.5's context window is 200K tokens, materially smaller than the 1M-token window on Fable 5.1, Opus 5, and Sonnet 5, and window size is unrelated to whether an effort parameter exists. Option C invents a shared surcharge; Haiku 4.5 is priced at $1/$5 per MTok input/output and Opus 5 at $5/$25, distinct price points with no documented equivalence for raising effort.

Answer: A
Claude Code subagents let a lead agent spawn multiple agents to work on different parts of a task in parallel and merge the results, useful for isolated, independent workstreams, which is exactly three unrelated modules: option A. Option B's hooks run an automatic step such as linting around edits; they enforce consistency, they do not split independent work across parallel agents. Option C's CLAUDE.md sets standing project context read at the start of a session; describing all three modules there does not make one sequential agent work them any faster. Option D's Routines are scheduled, recurring cloud runs such as a weekly dependency audit, not a mechanism for parallelizing a one-time refactor the engineer wants finished now.

Answer: B
Skills are reusable, packaged workflows, such as a shared /deploy-staging command, that let a team define a process once and reuse it consistently instead of re-describing it each time: option B. Option A's hook enforces an automatic, non-negotiable step around every commit regardless of intent, which is a different problem from giving engineers one shared, invocable deploy command they choose to run. Option C keeps the process as prose in a file read at session start, still asking each engineer to phrase a request in a way that reaches it, exactly the inconsistency the team wants to eliminate. Option D's subagents are for isolated, independent parallel workstreams, not for packaging one shared, repeatedly invoked deployment procedure.

Answer: C
Hooks run shell commands before or after Claude Code actions, such as auto-formatting after every edit or running lint before a commit, and are the way to enforce an automatic, non-negotiable step distinct from an instruction the model could skip: option C. Option A's CLAUDE.md sets standards and context Claude reads at session start, but it remains an instruction the model could still skip under a competing priority, not a guarantee. Option B's auto memory saves learnings across sessions as Claude works; it is not a mechanism for compelling a specific command to run before a specific action. Option D's subagents split independent work into parallel workstreams; they do not gate the very commit already in progress on a required check.

Answer: D
The Model Context Protocol is an open standard for connecting Claude to external tools and data sources such as Google Drive, Jira, and Slack without custom integration code per system, which is exactly what this workflow needs: option D. Option A's uploaded exports are static and one-directional, stale as soon as a ticket updates, and unable to post a message back into Slack. Option B's Structured Outputs shapes the JSON schema of Claude's own response; it has no mechanism to read a live Jira ticket or write into a Slack channel. Option C's chain-of-thought prompting organizes Claude's reasoning in labeled sections of a single reply; it does not connect to either external system or move any data between them.

Answer: A
Structured Outputs constrains Claude's response to conform to a given JSON schema and is the modern, reliable way to get machine-parseable output, explicitly replacing the older trick of prefilling a partial response like an opening brace; that is option A, and it is also Anthropic's stated migration path now that prefilled last-turn messages return an error on 4.6-generation-and-later models. Option B's longer history is not a documented fix and does not constrain the final response to a schema, so malformed output can still slip through. Option C's Cowork is a working session that acts on a user's own files and connected apps; it has no role in a backend service's own JSON-parsing request-response pipeline. Option D's tagged reasoning separates thinking from a final answer in prose, but the answer block itself is not schema-constrained and can still fail to parse as valid JSON.

Answer: A
Diagnosis starts by making the failure observable and changing one thing at a time; that is option A. A report that worked and now does not usually reflects a changed input, source document, or instruction, and the comparison finds which. Option B discards the evidence, and a regenerated prompt may reintroduce the same fault. Option C changes three variables at once, so even a fix teaches nothing about the cause. Option D fixes on one suspect first; a model update is one possibility among several, and the cheaper checks come earlier.

Answer: C
The observed fault is structural completeness, so the targeted fix names the sections and verifies them: option C. Anthropic's prompting guidance is to control the format explicitly and state what a complete output contains. Option B changes the sources, which affect content accuracy, not whether a section appears. Option D adds reasoning depth and creative license to a problem of omission; more thinking does not add a section nobody asked for. Option A removes the only structural cue and makes omissions more likely, not fewer.

Answer: B
Optimization removes repeated low-value work while keeping quality controls. Reformatting every week is a stable, describable step, so the fix is to put the template in the instructions and check conformance: option B. Option C trades a quality control for time and does not touch the actual bottleneck. Option D increases the volume reviewers must reshape. Option A spreads the manual work across more people instead of eliminating it, raising cost for the same output.

Answer: D
A durable improvement lives in the maintained configuration, not in one conversation: option D puts it in the shared project instructions or template, records why, and checks it on representative inputs before everyone relies on it. Option A strands the fix in personal history; memory is per user and per project, so colleagues never receive it. Option B removes the yardstick that showed the fix worked, so future regressions go unnoticed. Option C generalizes from one success and skips the testing that distinguishes a fix from a lucky run.

Answer: D
Domain 4 covers packaging a workflow so it runs the same way across a team instead of depending on individual habit. A Skill is a reusable, packaged workflow a team can share so a process is defined once and reused consistently, which is option D. Option B keeps the exact inconsistency the team is trying to fix, since letting each recruiter customize restates the current failure as a plan. Option A copies conversation history, which is personal and not a shareable, invocable process definition. Option C stores the prompt in one person's private project, so the rest of the team still depends on that one recruiter being reachable rather than on a workflow everyone can run directly.

Answer: A
Domain 4 covers connecting Claude Code to the external systems a workflow actually depends on. The Model Context Protocol lets Claude Code connect to servers such as a ticketing system, so it can read and write Jira and a deployment board directly as part of the pipeline, which is option A. Option B keeps the manual copying the engineer wants to remove, just moved into the chat window. Option C has Claude infer status instead of reading the source system, producing a guess where a live lookup exists. Option D uses a hook for the wrong purpose; a hook enforces a fixed script action like reformatting text, it does not give Claude read or write access to an external ticketing or tracking system.

Answer: B
Domain 4 covers matching the integration surface to a multi-file, multi-app working session while keeping the required human control. Claude Cowork works directly in a user's files, folders, and connected apps such as a shared drive and a signing tool such as DocuSign, which fits dozens of contracts and an external signing step; keeping attorney sign-off before sending preserves the needed check, which is option B. Option A drops the connectors the team explicitly wants, back to one file at a time with manual copying. Option C uses an artifact, a side panel for a single self-contained document, not a mechanism for working across many drive files and an external signing system. Option D removes the attorney's sign-off and lets Claude finalize contracts, exactly the control the team wants to keep.

Answer: C
Domain 4 places a control at the step where an error would cause harm, sized to the actual risk. The purchase order is submitted through a client tool the team's own application executes, and that call is exactly where a review gate can be inserted; a mis-sized order above the threshold has already caused a warehousing problem, so option C puts the required review there while letting routine small orders move faster. Option B reviews the low-stakes, high-volume orders and leaves the costly ones unchecked, inverting the risk. Option A reviews everything at the same rate, spending scarce reviewer time on orders that rarely need a change. Option D treats a successful system receipt as proof the quantity and price were correct, but a confirmation only means the order arrived, not that its size was right.

Answer: D
Domain 4 starts with requirements and use-case analysis: map the process before changing it. Option D names what has to be known to decide where Claude assists and where a person stays accountable. Option A designs the output before the process; wording is downstream of decision points and owners. Option B is a configuration choice that presupposes the steps are already defined; picking Manual, Auto, or Skip approval before the map exists is premature. Option C schedules and tools a launch for a process nobody has yet examined.

Answer: D
Domain 4 places a control where an error would do harm. Misassigned owners create missed work, so option D inserts a focused check at the distribution step while keeping the drafting gain. Option A accepts a known error rate on the field that matters most. Option B removes the useful content instead of correcting it; a summary without owners does not drive action. Option C asks for a promise the model cannot keep; an instruction changes phrasing, not the model's ability to know who agreed to what in the room.

Answer: C
Domain 4 asks associates to communicate Claude's value and its limits. Option C names the task it speeds up, the benefit, and the controls that stay with people, which a stakeholder can hold the team to. Option B overstates: Research citations make claims easier to check; they do not make them true. Option D inverts the dependency; a Project applies a brief, it does not discover one, and vague requirements produce vague synthesis. Option A promises replacement of the reviewers who catch errors, which misrepresents both the tool and responsible use.

Answer: A
Domain 4 measures workflow value end to end. A faster draft that needs more fixing may cost more overall, so option A looks at total cycle time, classifies the errors, and fixes the step that produces them before more teams inherit it. Option B scales a net loss because one local metric improved. Option C moves the cost off the books; correction time is part of the same workflow. Option D stops learning at the moment the data becomes useful, so the weak step would never be identified.

Answer: D
Domain 4 asks where to place a control that must never be skipped. A hook runs a shell command before or after a defined action, so the logging step executes outside the model's discretion every time; that is option D. Option A is an instruction, and an instruction is a request the model can still miss under load, which is exactly the failure the lead wants to close. Option B adds reasoning depth, not a guaranteed action; thinking harder about whether to log is not the same as logging. Option C keeps the default auto trigger, which lets Claude decide per ticket whether the tool is worth calling, the opposite of a non-negotiable step.

Answer: D
Domain 4 places a control where an error would cause harm, not evenly across every step. Payout approval is a finance decision, one of the Usage Policy's High-Risk Use Cases requiring a qualified professional to review it before it is finalized, so option D puts the required gate there and lets the low-risk draft move faster. Option C spends review effort on the draft explanation at the same rate as the payout, slowing the safe step without reducing the real risk. Option A puts the gate on the wrong step; a wording issue in a draft is cheap to fix, while an unreviewed incorrect payout has already left the business. Option B removes the review after a clean streak, but a clean history does not bound the size of the next error.

Answer: A
Domain 4 covers using Claude Code's coordination features to fit the shape of the work. Forty independent, self-contained changes are exactly the isolated workstream case subagents are built for: a lead agent assigns and merges parallel work, which is option A. Option B treats independent work as if it were sequential, so the team pays full serial time for no dependency reason. Option D substitutes a single scripted hook for judgment on forty distinct codebases, and a hook enforces a fixed step, not variable code migration. Option C confuses a connector to an external system with parallel execution capacity; MCP grants access to read and write a system, it does not split or coordinate the work across services.

Answer: C
Domain 4 covers integrating Claude into workflows that must run on a fixed cadence without a person present to trigger them. A routine is a scheduled, recurring Claude Code run in the cloud that keeps executing even when the user's machine and the analyst are both offline, which is option C. Option A depends on the analyst's memory and availability, exactly the failure vacation weeks expose. Option B asks Claude to remember and self-initiate a task, but the model does not wake itself on a calendar; something still has to trigger it. Option D is a working session a person opens manually, so it inherits the same single point of failure as option A.

Answer: D
Domain 4 covers wiring Claude into an existing CI/CD pipeline so a step runs automatically for every event, not only when someone remembers to start it. GitHub Actions and GitLab CI/CD let Claude Code automate PR review and issue triage as part of the pipeline itself, which is option D. Option B still depends on a person remembering to invoke a skill, the same trigger problem the team is trying to remove. Option A schedules a person's time, not an automatic per-pull-request check, and a week is too coarse for a step meant to run before a human opens the request. Option C is manual pasting with no trigger at all, so pull requests without an available reviewer get no review.