Domain 1: Agentic Architecture & Orchestration
27% of the exam.
- 1.1Agentic loopsBuild the loop that sends the conversation to Claude, runs the tools it requests, appends the results, and stops only when the stop reason is end_turn.
- 1.2Coordinator-subagent orchestrationUse one coordinator to decompose a task, pick and delegate to specialist subagents, route all communication, and refine synthesis until coverage is sufficient.
- 1.3Subagent invocation and context passingSpawn subagents with the Task tool, define each type with an AgentDefinition, and pass every needed finding explicitly as structured data because subagents start with no context.
- 1.4Workflow enforcement and handoffEnforce must-always-happen steps with hooks or prerequisite gates in code, decompose multi-concern requests, and hand off to humans with a structured summary.
- 1.5Agent SDK hooksUse PostToolUse hooks to normalize tool results and interception hooks to block policy-violating calls, choosing hooks over prompts when compliance must be guaranteed.
- 1.6Task decomposition strategiesChoose prompt chaining for predictable multi-aspect work and adaptive decomposition for open-ended investigation, splitting large reviews into focused passes.
- 1.7Session state, resumption and forkingResume named sessions, fork a shared analysis to compare approaches, flag changed files on resume, and start fresh with a summary when prior results are stale.
Domain 2: Tool Design & MCP Integration
18% of the exam.
- 2.1Tool interfaces and descriptionsWrite tool descriptions that state purpose, inputs, examples, edge cases, and boundaries, and rename or split tools so Claude can tell similar ones apart.
- 2.2Structured error responsesReturn tool failures with the isError flag, an error category, a retryable flag, and a readable message so the agent can retry, fix input, explain policy, or escalate.
- 2.3Tool distribution and tool_choiceGive each agent only the tools its role needs, swap generic tools for constrained ones, and use tool_choice to require or force a tool call.
- 2.4MCP servers in Claude Code and agentsConfigure shared servers in .mcp.json and personal ones in ~/.claude.json, reference secrets through environment variables, and expose catalogs as resources.
- 2.5Built-in toolsPick Grep for file contents, Glob for file paths, Read and Write for whole files, and Edit for unique-text changes, exploring codebases incrementally from entry points.
Domain 3: Claude Code Configuration & Workflows
20% of the exam.
- 3.1CLAUDE.md hierarchy and organizationPlace always-loaded instructions at the user, project or directory level, and keep them modular with @import and .claude/rules/ so each session loads only what applies.
- 3.2Slash commands and skillsPackage repeatable workflows as project or personal commands and skills, using SKILL.md frontmatter to isolate output, restrict tools and prompt for arguments.
- 3.3Path-specific rulesAttach conventions to glob patterns in .claude/rules/ so they load only when matching files are edited, wherever those files live.
- 3.4Plan mode vs direct executionPlan first when a task is large, multi-file or has several valid approaches; execute directly for small, clear changes; use the Explore subagent to keep discovery out of the main context.
- 3.5Iterative refinementSteer Claude with concrete input/output examples, tests written first, and the interview pattern, reporting interacting issues together and independent ones one at a time.
- 3.6CI/CD integrationRun Claude Code non-interactively with -p, request structured JSON output for the pipeline, supply context through CLAUDE.md, and review with an independent instance.
Domain 4: Prompt Engineering & Structured Output
20% of the exam.
- 4.1Explicit criteria in promptsReplace vague review instructions with explicit report-versus-skip criteria and severity examples, disabling noisy categories temporarily to protect developer trust.
- 4.2Few-shot promptingUse two to four targeted examples to lock in output format, show how to handle ambiguous cases, and reduce hallucination across varied document structures.
- 4.3Structured output with tool useDefine a tool whose input schema is the shape you want, pick the right tool_choice, and design nullable fields and extensible enums so the model reports rather than invents.
- 4.4Validation, retry and feedback loopsValidate extractions, retry with the document and specific errors, recognize when information is simply absent, and add fields that make failures analyzable.
- 4.5Batch processingUse the Message Batches API for latency-tolerant work at half the cost, schedule submissions against the SLA, and resubmit only failures by custom_id.
- 4.6Multi-instance and multi-pass reviewReview generated code with a fresh instance rather than the one that wrote it, and split large reviews into per-file passes plus a cross-file integration pass.
Domain 5: Context Management & Reliability
15% of the exam.
- 5.1Conversation context managementKeep exact facts alive in long conversations with a protected case facts block, trimmed tool output and position-aware ordering.
- 5.2Escalation and ambiguity resolutionEscalate on explicit requests, policy gaps and lack of progress, not on mood or self-reported confidence, and ask for identifiers on multi-match lookups.
- 5.3Error propagation across agentsSubagents retry transient failures locally and return structured error context (type, attempt, partials, alternatives) so the coordinator can recover and the report shows gaps.
- 5.4Context in large codebasesCounter context degradation with scratchpad files, subagent delegation, phase summaries, /compact and manifest-based crash recovery.
- 5.5Human review and calibrationMeasure accuracy by segment, calibrate field-level confidence on labeled data, route low-confidence cases to people and keep sampling the automated rest.
- 5.6Provenance and uncertaintyCarry claim-source mappings and dates through synthesis, annotate conflicts instead of picking one, and separate established from contested findings.