Validates that you can design Claude solutions end to end: agentic architecture, tool and MCP design, Claude Code workflows, prompt engineering, and context management. Scenario-based exam.
Who it's for: Solution architects and engineers designing Claude deployments, choosing platforms, and defining integration patterns.
Domains and weights from the official Anthropic exam guide (v1.0). Your score report shows percent-correct per domain.
Design and implement agentic loops, multi-agent systems, and orchestration patterns with the Claude Agent SDK.
Design effective MCP tool interfaces, implement structured error responses, and distribute tools across agents.
Configure Claude Code with CLAUDE.md files, Agent Skills, MCP integrations, and CI/CD workflow automation.
Engineer prompts for reliable structured output using JSON schemas, few-shot examples, and extraction patterns.
Manage context windows across long documents, multi-turn conversations, and multi-agent handoffs reliably.
Same style and cognitive level as the real exam. Click an answer to see the rationale.
A coordinator agent delegates research tasks to specialized subagents. A subagent needs information gathered earlier in the coordinator's conversation. How does the subagent get that context?