71 Claude certification concepts, made scannable.
Concepts are the 71 technical primitives the CCA-F (Claude Certified Architect - Foundations) exam tests on. Each page lowers friction with TLDR stats, diagrams, evidence pips, exam traps, and a sticky study rail. Grouped by exam domain. New here? See how this fits, or jump to the Claude Certified Architect certification guide.
Featured concept
Agentic loops: think, act, observe, decide. The mascot becomes the concept anchor.
D1 · 27% of the exam.
8 concepts in this domain.
Agentic Loops
An agentic loop turns one API response into a controlled, multi-turn system. The harness reads stop_reason, di…
Subagents
Subagents are specialized, isolated agents spawned by a coordinator to handle domain-specific tasks while pres…
Stop Reason
stop_reason is the authoritative struct field that says why Claude stopped, either end_turn (finished) or tool…
Session State & Persistence
Session state is the running message list (and any external persistence) that gives an agentic loop continuity…
Human-in-the-Loop Escalation
Escalation is the deterministic handoff path when policy thresholds, low confidence, or repeated failure condi…
Subagent State Handoff
Subagents hand off state three ways: structured messages (default - typed JSON envelopes through the orchestra…
Agent Instruction Files (CLAUDE.md / AGENTS.md)
CLAUDE.md is Claude-specific; AGENTS.md is the emerging cross-tool convention (Codex, Cursor, Aider). Portabil…
Task Decomposition Strategies
Task decomposition is the up-front architectural choice between two regimes: **fixed decomposition** (subtasks…
D2 · 18% of the exam.
6 concepts in this domain.
Tool Calling
Tool calling is how Claude decides to invoke external functions and pass structured arguments. Tools are routi…
Tool Choice
tool_choice is the parameter that controls whether Claude can decide to use a tool ("auto"), must call any too…
Model Context Protocol
MCP is a communication standard that lets Claude access pre-built tools, resources, and prompts from specializ…
SDK Hooks (Pre/PostToolUse)
Hooks are deterministic code that runs before or after tool calls. They enforce policy that prompt-only patter…
Tool Evaluation & Testing
Evaluation is how you measure tool-call correctness against ground-truth datasets. A full deep-dive guide is c…
Built-In Tools vs Custom Tools vs Skills vs MCP
Built-in tools, custom tools, Agent Skills, and MCP servers are ==complementary layers==, not a power ranking …
D3 · 20% of the exam.
4 concepts in this domain.
CLAUDE.md Hierarchy
CLAUDE.md is persistent project memory loaded automatically by Claude Code. A three-level hierarchy (user → pr…
Plan Mode
Plan Mode forces Claude Code to draft and approve a multi-step plan before executing. Use it for non-trivial c…
Skills
Skills are reusable, on-demand task workflows defined in markdown with YAML frontmatter. Unlike CLAUDE.md (alw…
Checkpoints & Session Management
Checkpoints save and restore conversation state. A full deep-dive guide is coming soon. {🟡 Claude Code checkp…
D4 · 20% of the exam.
10 concepts in this domain.
System Prompts & Instructions
System prompts establish role, constraints, format, and tool guidance. Anatomy: role, capability boundaries, s…
Prompt Caching
Prompt caching reduces cost (~90%) on repeated context like long system prompts and tool definitions. Cache br…
Batch API
Message Batches API: 50% discount for async, non-time-sensitive workloads. {🟢 Anthropic batch API}…
Structured Outputs
Structured outputs guarantee Claude returns JSON matching a schema instead of natural language. Force a tool w…
Vision & Multimodal
Vision lets Claude process images alongside text. A full deep-dive guide is coming soon. {🟢 messages content …
Streaming
Streaming returns tokens as they're generated for low-latency UX. A full deep-dive guide is coming soon. {🟢 m…
Attention Engineering
Attention engineering is the discipline of placing critical context where the model attends most strongly, hig…
4D Framework
Delegation, Description, Discernment, Diligence, Anthropic's 4D framework for agent prompt design. A full deep…
Prompt Engineering Techniques
Prompt engineering is a craft of seven techniques that turn a brittle one-shot prompt into a production-grade …
Multi-Instance & Multi-Pass Review Architectures
A multi-pass review architecture runs more than one independent Claude pass over the same artifact and reconci…
D5 · 15% of the exam.
2 concepts in this domain.
CCA-P concepts.
Concepts tested by the Claude Certified Architect - Professional exam.
D1 · 17% of CCA-P.
6 concepts in this domain.
Subagents
Subagents are specialized, isolated agents spawned by a coordinator to handle domain-specific tasks while pres…
Task Decomposition Strategies
Task decomposition is the up-front architectural choice between two regimes: **fixed decomposition** (subtasks…
Translating Business Requirements into Claude Solutions
Translating a business problem into a Claude solution means producing a build-ready ==requirements spec==, not…
End-to-End Architecture Design (Input to Feedback Loop)
An architect-level design task is never just "the prompt in the middle" - it is the full pipeline: how input a…
Workflow vs Agent: Choosing the Right Architecture
Pick the cheapest tier that actually satisfies the task: an ==augmented LLM== call for a single unaided prompt…
Aligning Solutions to Business Value Pillars
Every architectural choice - pattern, model tier, guardrail overhead - has to trace back to a named business-v…
D2 · 13% of CCA-P.
5 concepts in this domain.
System Prompts & Instructions
System prompts establish role, constraints, format, and tool guidance. Anatomy: role, capability boundaries, s…
Prompt Caching
Prompt caching reduces cost (~90%) on repeated context like long system prompts and tool definitions. Cache br…
Prompt Engineering Techniques
Prompt engineering is a craft of seven techniques that turn a brittle one-shot prompt into a production-grade …
Context Window Management
Context window management is how you keep long conversations within limits without dropping critical facts. Pa…
Claude Model Selection & Tradeoffs
Match the model tier to the task, not the other way round: ==Haiku== for high-volume, straightforward work at …
D3 · 19% of CCA-P.
5 concepts in this domain.
Model Context Protocol
MCP is a communication standard that lets Claude access pre-built tools, resources, and prompts from specializ…
Authentication & Access Control for Agent Systems
Agent authentication and authorization is a stack of controls, not one setting: which identity the process run…
Accuracy-Latency-Cost Tradeoff Analysis
Every production Claude deployment sits on a capability/speed/cost triangle, and the exam trap is treating it …
Observability & Monitoring for Agents at Scale
Observability for a Claude agent fleet spans layers you must monitor separately: platform-level usage/cost acc…
Progressive Discovery vs Monolithic Context Strategy
As an agent's tool/skill/knowledge surface grows, an architect chooses between loading everything upfront (mon…
D4 · 16% of CCA-P.
4 concepts in this domain.
Tool Evaluation & Testing
Evaluation is how you measure tool-call correctness against ground-truth datasets. A full deep-dive guide is c…
Accuracy-Latency-Cost Tradeoff Analysis
Every production Claude deployment sits on a capability/speed/cost triangle, and the exam trap is treating it …
Observability & Monitoring for Agents at Scale
Observability for a Claude agent fleet spans layers you must monitor separately: platform-level usage/cost acc…
Debugging LLM System Failures
A Claude application fails in two structurally different places: the ==integration layer== (the call never rea…
D5 · 14% of CCA-P.
5 concepts in this domain.
Human-in-the-Loop Escalation
Escalation is the deterministic handoff path when policy thresholds, low confidence, or repeated failure condi…
Guardrails & Safe Deployment Practices
Guardrails are not one control but a ==layered system==: model-level training behavior, input-level content sc…
LLM System Risks & Failure Modes
Architects must reason about failure modes that don't exist in deterministic software: non-determinism, prompt…
Regulatory Compliance for AI Systems (GDPR, HIPAA, FedRAMP)
Compliance is a match-the-surface-to-the-regime exercise: an architect picks the deployment surface (API direc…
Ethical AI Considerations (Bias, Fairness, Transparency)
Ethical AI for an architect is not a vendor guarantee, it is three published artifacts to consult: ==Claude's …
D6 · 14% of CCA-P.
5 concepts in this domain.
Structured Stakeholder Discovery
Discovery for a Claude-based solution has to surface things traditional SDLC discovery skips: whether the task…
Communicating AI Decisions & Value to Stakeholders
An architect's technical trade-offs (model tier, workflow vs agent, guardrail depth) are business decisions in…
Stakeholder Feedback Loops & SLA Alignment
A Claude-based system's "SLA" is not one number - an architect has to keep at least three infrastructure-side …
Architecture Documentation & Implementation Handoff
Handoff for a Claude-based architecture needs everything a traditional diagram needs (components, data flow, i…
AI Solution Lifecycle Management
Deployment isn't the finish line - an architect owns a continuous loop of ==discovery, design, handoff, monito…
CCD-F concepts.
Concepts tested by the Claude Certified Developer - Foundations exam.
D1 · 14.7% of CCD-F.
3 concepts in this domain.
Workflow vs Agent: Choosing the Right Architecture
Pick the cheapest tier that actually satisfies the task: an ==augmented LLM== call for a single unaided prompt…
Building Agents: Client SDK vs Agent SDK vs Managed Agents
Anthropic gives developers three layers for building a Claude agent, and they are NOT interchangeable names fo…
Agent Design Patterns: Memory, Context Management & Third-Party Frameworks
An agent's execution mechanism is the ==tool-use loop==: call a tool, read real environmental feedback ("groun…
D2 · 33.1% of CCD-F.
10 concepts in this domain.
Tool Calling
Tool calling is how Claude decides to invoke external functions and pass structured arguments. Tools are routi…
CLAUDE.md Hierarchy
CLAUDE.md is persistent project memory loaded automatically by Claude Code. A three-level hierarchy (user → pr…
Prompt Caching
Prompt caching reduces cost (~90%) on repeated context like long system prompts and tool definitions. Cache br…
Batch API
Message Batches API: 50% discount for async, non-time-sensitive workloads. {🟢 Anthropic batch API}…
Vision & Multimodal
Vision lets Claude process images alongside text. A full deep-dive guide is coming soon. {🟢 messages content …
Streaming
Streaming returns tokens as they're generated for low-latency UX. A full deep-dive guide is coming soon. {🟢 m…
Translating Business Requirements into Claude Solutions
Translating a business problem into a Claude solution means producing a build-ready ==requirements spec==, not…
Systems Life Cycle Management for Claude Applications
Systems life cycle management is a general IT discipline - plan, develop, test, deploy, operate, maintain, ret…
Software Engineering Foundations for Claude Applications
The Claude Messages API ==is a plain REST API== that accepts and returns JSON - `POST /v1/messages` with a sta…
Claude Application Design Across Interfaces
Claude is one model family reached through several host surfaces, claude.ai/Desktop, Claude Code, and the raw …
D3 · 3.1% of CCD-F.
1 concept in this domain.
D4 · 2.6% of CCD-F.
1 concept in this domain.
D5 · 16.8% of CCD-F.
4 concepts in this domain.
Prompt Caching
Prompt caching reduces cost (~90%) on repeated context like long system prompts and tool definitions. Cache br…
Claude Model Selection & Tradeoffs
Match the model tier to the task, not the other way round: ==Haiku== for high-volume, straightforward work at …
LLM Fundamentals: Tokens, Sampling & Non-Determinism
Claude is autoregressive: it predicts one token at a time from a probability distribution conditioned on every…
Technical Fundamentals: SDKs, REST, and WebSockets
For standard model inference, the ==Messages API== (`POST /v1/messages`) is the primary endpoint, tools, struc…
D6 · 11% of CCD-F.
4 concepts in this domain.
Structured Outputs
Structured outputs guarantee Claude returns JSON matching a schema instead of natural language. Force a tool w…
Attention Engineering
Attention engineering is the discipline of placing critical context where the model attends most strongly, hig…
Prompt Engineering Techniques
Prompt engineering is a craft of seven techniques that turn a brittle one-shot prompt into a production-grade …
Context Window Management
Context window management is how you keep long conversations within limits without dropping critical facts. Pa…
D7 · 8.1% of CCD-F.
4 concepts in this domain.
SDK Hooks (Pre/PostToolUse)
Hooks are deterministic code that runs before or after tool calls. They enforce policy that prompt-only patter…
Authentication & Access Control for Agent Systems
Agent authentication and authorization is a stack of controls, not one setting: which identity the process run…
Guardrails & Safe Deployment Practices
Guardrails are not one control but a ==layered system==: model-level training behavior, input-level content sc…
AI Application Security: Prompt Injection & Untrusted Input
Prompt injection splits into two threat models with different defenses: ==direct injection / jailbreaks== (you…
D8 · 10.6% of CCD-F.
3 concepts in this domain.
Tool Calling
Tool calling is how Claude decides to invoke external functions and pass structured arguments. Tools are routi…
Model Context Protocol
MCP is a communication standard that lets Claude access pre-built tools, resources, and prompts from specializ…
Built-In Tools vs Custom Tools vs Skills vs MCP
Built-in tools, custom tools, Agent Skills, and MCP servers are ==complementary layers==, not a power ranking …
CCA-A concepts.
Concepts tested by the Claude Certified Associate - Foundations exam.
D1 · 14% of CCA-A.
3 concepts in this domain.
Prompt Engineering Techniques
Prompt engineering is a craft of seven techniques that turn a brittle one-shot prompt into a production-grade …
Task Decomposition Strategies
Task decomposition is the up-front architectural choice between two regimes: **fixed decomposition** (subtasks…
Prompting Strategy by Task Type: Analysis, Research, Drafting, Brainstorming
"Be clear and direct" is Anthropic's base rule for every Claude prompt, but ==how you apply it changes with th…
D2 · 21% of CCA-A.
5 concepts in this domain.
Human-in-the-Loop Escalation
Escalation is the deterministic handoff path when policy thresholds, low confidence, or repeated failure condi…
Evaluating Claude Output for Accuracy & Completeness
Accuracy and completeness are ==two independent checks==, not one: an output can be fully correct on every cla…
Identifying Hallucinations, Inconsistencies & Bias in AI Output
Hallucination is Anthropic's own term for Claude generating text that's factually incorrect or inconsistent wi…
Editing & Adapting Outputs for the Intended Audience
A first Claude draft is rarely the version you hand to its actual reader. Adapting output for audience runs on…
Choosing Output Formats: Artifacts vs Inline vs Structured Data
Choosing how a Claude response looks is really two **independent** decisions, not one three-way pick. ==Axis 1…
D3 · 12% of CCA-A.
3 concepts in this domain.
Context Window Management
Context window management is how you keep long conversations within limits without dropping critical facts. Pa…
Claude Model Selection & Tradeoffs
Match the model tier to the task, not the other way round: ==Haiku== for high-volume, straightforward work at …
Selecting the Right Claude Product Feature
Claude.ai gives a business user four distinct surfaces, plain **chat**, **Projects**, **Research**, and **Arti…
D4 · 16% of CCA-A.
4 concepts in this domain.
Translating Business Requirements into Claude Solutions
Translating a business problem into a Claude solution means producing a build-ready ==requirements spec==, not…
Communicating AI Decisions & Value to Stakeholders
An architect's technical trade-offs (model tier, workflow vs agent, guardrail depth) are business decisions in…
Using Claude for Research, Planning & Solution Development
Beyond one-off Q&A, Claude is used as a research and planning partner across a three-stage arc: research a pro…
Integrating Claude into Existing Workflows
Integrating Claude into a workflow is a choice between ==augment== (a human still owns the process, Claude dra…
D5 · 12% of CCA-A.
3 concepts in this domain.
System Prompts & Instructions
System prompts establish role, constraints, format, and tool guidance. Anatomy: role, capability boundaries, s…
Configuring & Maintaining Claude Projects
A Claude Project is a self-contained workspace with its own chat history and its own knowledge base, configure…
Managing Uploaded Knowledge & Connectors
A Project's knowledge splits into two mechanisms that behave differently: **uploads** are a static snapshot a …
D6 · 15% of CCA-A.
4 concepts in this domain.
Regulatory Compliance for AI Systems (GDPR, HIPAA, FedRAMP)
Compliance is a match-the-surface-to-the-regime exercise: an architect picks the deployment surface (API direc…
Ethical AI Considerations (Bias, Fairness, Transparency)
Ethical AI for an architect is not a vendor guarantee, it is three published artifacts to consult: ==Claude's …
Identifying Appropriate vs Inappropriate AI Use Cases
Before building a workflow around Claude, judge whether a task is one Claude is ==allowed== and ==well-suited=…
Following Organizational AI Governance Policy
Following your organization's AI governance is an employee-level checklist run before acting: is this data cla…
D7 · 10% of CCA-A.
3 concepts in this domain.
Debugging LLM System Failures
A Claude application fails in two structurally different places: the ==integration layer== (the call never rea…
Adjusting Approach Based on Feedback & Results
When a Claude output misses the mark, the fix Anthropic teaches is diagnostic, not a re-roll: define what a go…
Optimizing AI Workflows for Efficiency
Optimizing a Claude workflow for efficiency AND effectiveness spans more than one lever: tightening the prompt…
6 questions about the concept catalog
Every Q is phrased as a real Google search query. Answers cite the same evidence-tagged sources used elsewhere on the site.