Anthropic · CCDV-F
Claude Certified Developer – Foundations (CCDV-F) Practice Tests
Validates foundational knowledge and skills for developers building applications with Anthropic's Claude AI models. Covers API integration, prompt engineering, model selection, agents, workflows, and security best practices.
Exam Details
Practice Questions
302
≈ 4 practice exams
Duration
120 minutes
Passing Score
720/1000
Difficulty
FoundationalLast Updated
Sep 2026
Topics Covered
Applications & IntegrationModel Selection & OptimizationAgents & WorkflowsPrompt & Context EngineeringTools & MCPsSecurity & SafetyClaude CodeResponsible AI Development
Claude Certified Developer – Foundations (CCDV-F) Practice Exam Preparation
Use this CCDV-F practice exam to prepare for Claude Certified Developer – Foundations (CCDV-F) with realistic questions, detailed explanations, and focused study modes. The practice bank includes 302 questions for Anthropic CCDV-F, so you can review the exam steadily instead of relying on one long cram session.
As you practice, pay extra attention to recurring topics such as Applications & Integration, Model Selection & Optimization, Agents & Workflows, Prompt & Context Engineering, and Tools & MCPs. Start with short sessions to identify weak areas, then move into timed quizzes once your accuracy is consistent.
The explanations are written for candidates preparing for Anthropic certification questions and reviewing how the concepts apply in practice. Use the free preview first, then unlock the full question bank when you are ready to build a complete study routine.
Sample Questions
5 sample questions with answers and explanations. The full bank has 302 questions, enough for 4 full-length practice exams.
Preview — answers shown1. A developer is building a Claude-powered application that processes user queries and must also handle images. They need to ensure the model can reason about the images. Which capability should they use? (Select one!)
AExtended thinking
BBatch API
CVision
DPrompt caching
Explanation
Vision capabilities allow the model to analyze images, which is required for image reasoning. Extended thinking is for complex reasoning tasks, prompt caching is for cost/latency optimization, and the Batch API is for asynchronous workloads. Using Vision is the correct way to handle image inputs.
2. A developer is configuring an MCP server that will run locally on the same machine as the Claude application. The server needs to communicate with Claude via standard input and output. Which transport should be used? (Select one!)
ANetwork sockets
Bstdio
CHTTPS
DWebSocket
Explanation
The stdio transport is used for locally spawned MCP servers, where the server communicates with Claude through standard input and output. Network sockets are for remote servers that run on a different machine. HTTPS and WebSocket are not standard MCP transports; they are general network protocols.
3. A developer is configuring Claude Code for a project that requires the model to run shell commands and edit files. They need to restrict the model from accessing certain directories. Which file should they modify? (Select one!)
AREADME.md
Bpackage.json
CCLAUDE.md
Dsettings.json
Explanation
settings.json governs permissions and tooling in Claude Code. It is the correct place to restrict directory access. CLAUDE.md carries project instructions, not permissions. package.json and README.md are not used for Claude Code permissions.
4. A developer wants to run a Claude Code session that provides real-time output as the model generates responses, but still allows interactive input. Which Claude Code feature should the developer enable? (Select one!)
AAuto-mode
BStreaming mode
CHeadless mode
DSession management
Explanation
Streaming mode provides real-time output during interactive sessions, allowing the developer to see responses as they are generated. Headless mode is non-interactive, auto-mode involves agent-driven pauses, and session management handles conversation state. Streaming mode is the correct feature for real-time output with interactivity.
5. A developer is building an agent that can issue refunds to customers. The agent has access to a refund API. What is required before the agent executes a refund? (Select one!)
AA tool description that says 'be careful'
BA human approval gate
CA system prompt warning
DA max_tokens limit
Explanation
Human approval gates are required for actions with meaningful real-world consequences, such as issuing refunds. A system prompt warning, a tool description, or a max_tokens limit are not sufficient safeguards. The developer must implement a programmatic gate that pauses the agent for human approval before the refund API is called.
$17.99
One-time access to this exam
302 questions (4 practice exams' worth)
Unlimited timed exam simulations
Yours forever — no subscription
Detailed explanations
Free preview stays available