Looking for your active Pro access before showing Course Notes. This usually takes just a moment.
Course Notes preview
Unlock Pro for the full per-exam reference guide.
Preview one piece from each section. Pro includes every Course Notes section, summary, key point, common mistake, exam tip, related-question review, and PDF export.
Includes full Course Mode and Course Notes.
Section 1FundamentalsPreview
More in this section
5 more summary sections in Pro version
4 more key points in Pro version
31 more related questions in Pro version
Summary
Claude architecture starts with choosing the right surface for the job. The direct API gives an application full control over prompts, message history, tools, retries, storage, and user experience, while Managed Agents are a better fit when Anthropic-managed sessions and built-in autonomy reduce the amount of orchestration the product team must build.
Key Points
Multimodal Input: Use multimodal input when Claude must reason over images, screenshots, diagrams, scanned documents, or mixed text-and-visual evidence instead of relying on text alone.
Section 2API ArchitecturePreview
More in this section
6 more summary sections in Pro version
4 more key points in Pro version
20 more related questions in Pro version
Summary
Claude API design begins with authentication, workspace separation, and request sizing. Production systems should keep API keys out of client-side code, separate environments or use cases with workspaces where appropriate, and handle large prompts, files, and rate behavior before the first user-facing rollout.
Key Points
Messages API: Use the Messages API for new Claude applications that need role-based turns, system prompts, tools, streaming, vision, files, and stateless conversation control.
Section 3PromptingPreview
More in this section
6 more summary sections in Pro version
4 more key points in Pro version
20 more related questions in Pro version
Summary
Prompting architecture separates durable behavior from one-off user input. Put role, tone, safety constraints, and output contract in the system prompt, then keep the user turn focused on the specific task and source material for that request.
Key Points
System Prompt: Use a system prompt for durable behavior rules, role instructions, output requirements, and safety constraints that should apply across the whole request.
Section 4Tooling & AgentsPreview
More in this section
6 more summary sections in Pro version
4 more key points in Pro version
20 more related questions in Pro version
Summary
Tool architecture starts with a precise contract. Each tool needs a name, description, and input schema that tells Claude when the tool is appropriate and tells the application exactly what arguments it may receive.
Key Points
Tool Definition: A tool definition is the contract Claude reads to decide when a tool applies and the schema the application uses to validate the arguments Claude returns.
Section 5Model ContextPreview
More in this section
5 more summary sections in Pro version
4 more key points in Pro version
14 more related questions in Pro version
Summary
Model Context Protocol is an integration standard for connecting AI clients to external tools and data sources. In Claude Code, MCP is not a separate model or a mission-control layer; it is the protocol used to expose approved capabilities from MCP servers to an MCP client.
Key Points
Model Context Protocol (MCP): MCP is an open protocol for connecting AI clients to external tools, resources, and data sources through servers that expose approved capabilities.
Section 6Code WorkflowPreview
More in this section
5 more summary sections in Pro version
4 more key points in Pro version
9 more related questions in Pro version
Summary
Claude Code is most useful when the task needs repository context plus action. It can inspect files, explain unfamiliar modules, edit code, run commands, and use development tools, so the architect's decision is how much execution authority to give it for the current repository and risk level.
Key Points
Worktree: Use a worktree when parallel development needs separate working directories and branches so concurrent Claude Code sessions do not overwrite each other.
Section 7Security & GovernancePreview
More in this section
5 more summary sections in Pro version
4 more key points in Pro version
9 more related questions in Pro version
Summary
Least privilege should shape every Claude Code deployment. Give users and agents only the repository, command, tool, and external-system access needed for the task, then expand permissions deliberately when the risk is understood.
Key Points
Least Privilege: Use least privilege to limit Claude Code, users, and integrations to only the files, commands, tools, and external systems required for the current task.
Section 8Work & ProjectsPreview
More in this section
4 more summary sections in Pro version
3 more key points in Pro version
3 more related questions in Pro version
Summary
Claude Projects organize work around a durable context boundary. A project can hold instructions, knowledge, and shared materials for a product, client, team, or initiative so users do not have to rebuild the same context in every conversation.
Key Points
Claude Project: Use a Claude Project when a team or individual needs a persistent workspace with shared instructions, files, and context for related work.
Search catalog
Find a practice exam
Flexible search understands AI-901, ai901, ai 901, 901, ai, network plus, and saa c03.