Try one now
A batch worker runs parallel requests against Claude Sonnet 5. Under traffic spikes some requests come back HTTP 429 with an `error.type` of `rate_limit_error` and a `retry-after` header.
What should the client do with those requests?
Answer it and read the explanation →About the CCDV-F exam
- Length
- 120 minutes
- Questions
- 53 items
- Passing score
- 720 on a 1,000 scale
- Price
- $125 USD
- Question types
- Multiple choice and multiple response
Checked against the provider's certification page and exam guide on 2026-09-15. Providers change fees and formats; confirm before you book. Register for the official exam ↗
What the exam covers
120 minutes, 53 questions, and a 720 out of 1,000 passing score. Anthropic lists multiple-choice and multiple-response items.
- Agents and workflows15%
- Applications and integration33%
- Claude Code3%
- Evaluation, testing, and debugging3%
- Model selection and optimization17%
- Prompt and context engineering11%
- Security and safety8%
- Tools and MCP11%
Weights are from the provider's published exam guide.
Who this is for. Developers implementing applications with Claude APIs and tools.
How to study. Follow the complete request and response lifecycle. Practice tool-result handling, streaming events, and caching boundaries.
Our bank has 105 questions and uses single-answer multiple choice. Official exams may use additional formats. This is an independent practice set.
All 105 practice questions
Open a question to try it and read the reasoning. Related questions may also appear in other relevant tracks.
Claude Code 9
- Where should the personal preferences live?easy
- Where should the rules be placed?medium
- What is the recommended approach?easy
- Which permission rule accomplishes this?easy
- Which allowedTools value is correct?hard
- How should the server be configured?medium
- Which mechanism provides that guarantee?medium
- Which frontmatter setting achieves this?medium
- How should the helper be defined?medium
Applications & Integration 23
- What should the client do with those requests?medium
- What is the correct way to supply system instructions?easy
- Which server-sent event carries each increment of the reply?medium
- How should the client assemble the tool input before executing the call?medium
- What is the correct structure for an image content block in the Messages API?medium
- Which change prevents the duplicates?medium
- What should the developer do first?medium
- How should the run be submitted and reconciled?medium
- How should the PDF reach the model?medium
- How should the service handle this response?medium
- How should the client assemble them?medium
- How should the pipeline collect the translations?medium
- Why does the request fail validation?medium
- How should the application supply these documents?medium
- Which server-sent event carries the final output token count?medium
- How should the assistant turn be included in that request?medium
- Which approach gives the closest count before the request is sent?easy
- Which request configuration enforces that?easy
- Which practice would have caught the error early?medium
- What explains the failure?medium
- What should the team change to keep customer files isolated?hard
- What is the recommended fix?easy
- How should the schema be corrected?medium
Model Selection & Optimization 15
- How is the cache breakpoint declared in the request?medium
- What is the most likely cause?medium
- Which change restores prefix reuse?medium
- What does that measurement imply for the routing design?hard
- Which evidence directly establishes a cache read?medium
- Which model assignment fits both workloads?easy
- How is a cache read priced against base input tokens on this model?medium
- What explains the difference between the two models?medium
- How does the cache entry behave under this query pattern?medium
- Which approach gets the same quality for less?medium
- Which change is most likely to raise the cache hit rate?medium
- Which change raises effective throughput without a limit increase?hard
- What should the team do first?medium
- Which change restores cache reads on this conversation?hard
- Which source should the service read those limits from instead?easy
Tools & MCP 12
- What must the next request append to the conversation history?medium
- How should the platform team expose the capability?medium
- Which change best stops the agent from calling the wrong one?medium
- What should the application send back on the next request?medium
- Which part of a tool definition tells the host that a tool changes state?medium
- Which change stops those invalid values at the source?easy
- Which limitation of the MCP connector blocks this plan?medium
- How should the server report that failure?medium
- Which MCP primitive should the catalog pages use?medium
- Which change fixes the parse errors?easy
- How should the server tell connected clients the list has changed?easy
- Which change addresses both problems?medium
Evaluation, Testing & Debugging 9
- What is the best next step?medium
- What should the release decision rest on?medium
- Which evaluation method fits?medium
- Which option should the team choose?medium
- How should the grader be set up?medium
- Which detail identifies the individual failing calls for support?easy
- What is happening, and what fixes it?hard
- How should the criterion be rewritten?easy
- What change keeps extended reasoning working on the new model?medium
Security & Safety 10
- Which change removes both weaknesses at once?medium
- What is Anthropic's default handling of the inputs and outputs of those API requests?medium
- Which change best hardens the agent against instructions carried in those files?medium
- Which property of the code execution sandbox blocks this plan?medium
- Which setup matches Anthropic's precautions for running computer use?medium
- Which approach meets that requirement?easy
- How must the server collect the key under the MCP specification?hard
- Which control stops a rebound request from being served?medium
- How must the MCP server handle that incoming bearer token?hard
- What should the team switch the pipeline to?medium
Prompt & Context Engineering 13
- Which control prevents the refund even if every screening layer misses the injection?hard
- What should replace the prefill?medium
- Which strategy keeps the thread coherent as it grows?medium
- How should examples be added to the prompt?easy
- Which prompt change most directly addresses the tone and rigor?easy
- Which prompt structure most reliably separates the three kinds of content?easy
- What prompt change should the team make?medium
- Which prompt change most directly addresses this?medium
- How should the prompt be reordered?easy
- How should the instruction be delivered?hard
- Where should the developer place that instruction?medium
- What is the simplest prompt fix?easy
- What is the recommended first step?medium
Agents & Workflows 14
- Which architecture best fits this traffic?medium
- Which control stops the runaway loop?medium
- Which workflow pattern should the team build around this validator?medium
- Which guardrail belongs in front of delete_namespace?medium
- How should the results go back to Claude?hard
- Which change lets the coordinator finish the set?medium
- Which architecture best fits these inquiries?medium
- Which Anthropic offering matches this requirement?medium
- Which context editing configuration meets this need?medium
- How should the agent carry state between sessions?medium
- What should the application send next?hard
- Which multi-model pattern fits this workload?hard
- How should the application continue the turn?medium
- Which implementation should the team choose?medium
Common questions
What can I practice for Claude Certified Developer: Foundations?
105 independent single-answer practice questions covering selected topics: Agents and workflows, Applications and integration, Claude Code, Evaluation, testing, and debugging, Model selection and optimization, Prompt and context engineering, Security and safety, Tools and MCP. Each includes an explanation and source link.
Is this an official certification or exam dump?
No. AI CertArena writes independent practice scenarios. We are not affiliated with or endorsed by the exam providers, and we do not publish live exam questions.
Does my practice score predict an exam pass?
No. Your score is the percentage correct in this practice set, including skipped questions. Our 80% practice goal is a study target, not an official cut score or calibrated pass prediction.
How should I use this practice bank?
Try a short set, review the explanation for every miss, and repeat those questions later. Use the official exam guide to check topics that this starter bank does not yet cover.
Can I save and share my progress?
Sign in with Google to keep completed attempts and topic evaluations across devices. You can sign in after finishing a set and it still counts. You can also choose to publish a saved practice result and download an image to share. A practice badge is not a vendor credential.