认证概览
了解 Claude Certified Architect 能证明什么，以及它为何重要。
认证概览 是 CoddyKit 上的免费 Claude Architect 课时。 这是第 1 节课，共 4 节。 你可以在下方免费阅读本课时的完整内容 — 然后在浏览器中使用内置代码编辑器和全天候 AI 导师进行实践。 这是 Claude Architect 学习路径的一部分，你的进度在网页和 CoddyKit 应用中同步。 Claude Architect 课程共包含 4 节课。
本课时的部分内容尚未翻译，以英文显示。
Welcome, Future Architect
The Claude Certified Architect credential proves you can design real, production-grade systems on Claude — not just write a clever prompt. Here's the big picture.
Who It's For
This is an architect-level cert aimed at solution architects with 6+ months of production Claude experience. Newer to it? You can still pass by learning the production patterns here.
Exam Format at a Glance
The exam is scenario-based multiple choice: a realistic situation, 4 options, exactly 1 correct. You won't recite definitions — you'll choose the right design decision.
Scoring and the Pass Line
Your result is a scaled score from 100 to 1000, and you pass at 720. There's no penalty for guessing, so answer every question — a blank beats nothing.
score = 745
PASS = 720
if score >= PASS:
print("Certified Claude Architect \U0001F389")
else:
print(f"Keep studying: {PASS - score} points to go")How Scenarios Are Sampled
The cert is built around 8 canonical scenarios, but any exam shows you just 4 of them. Since you can't predict which, the smart move is to prepare for all 8.
The Five Domains
Questions come from five weighted domains: D1 Agents 27%, D2 Tools/MCP 18%, D3 Claude Code 20%, D4 Prompting 20%, D5 Context 15%. The weights are your study budget.
Domain 1 Carries the Most Weight
At 27%, Agent Architecture and Orchestration is the largest domain. It covers the agentic loop: send a request, check stop_reason, run tools, repeat until end_turn.
# The agentic loop — driven by stop_reason, never by parsing text
resp = client.messages.create(
model="claude-sonnet-4-5",
max_tokens=1024,
messages=messages,
tools=tools,
)
while resp.stop_reason == "tool_use":
messages.append({"role": "assistant", "content": resp.content})
messages.append({"role": "user", "content": run_tools(resp)})
resp = client.messages.create(
model="claude-sonnet-4-5",
max_tokens=1024,
messages=messages,
tools=tools,
)
# loop exits cleanly on stop_reason == "end_turn"What the Other Domains Prove
The other domains each certify a skill: D2 tool descriptions and MCP, D3 CLAUDE.md and workflows, D4 prompting and JSON Schema, D5 context and reliability.
It Certifies Judgment, Not Trivia
The exam rewards architectural judgment over trivia. Many questions hide an anti-pattern — like stopping an agent by parsing its text for 'done' instead of stop_reason.
# Anti-pattern (a tempting WRONG answer):
if "done" in resp.content[0].text.lower():
stop() # fragile, model-phrasing dependent
# Correct: terminate on the protocol signal
if resp.stop_reason == "end_turn":
stop() # deterministic, model-drivenGrounded in the Real API
Every scenario assumes you know the Messages API: the model keeps no state, so you resend the full history every turn, and you read the four stop reasons.
resp = client.messages.create(
model="claude-sonnet-4-5",
max_tokens=1024,
system="You are a precise solutions architect.",
messages=messages, # FULL history every turn — model is stateless
tools=tools,
tool_choice={"type": "auto"},
)
print(resp.stop_reason) # end_turn | tool_use | max_tokens | stop_sequenceWhy the Credential Matters
Passing proves you can turn Claude from a chat tool into dependable infrastructure — agents that loop, tools that route, systems that fail gracefully. Next: the 8 scenarios and 5 domains.
Quick Check
A teammate sees their certification report and panics. Which interpretation is correct?
Recap: The Cert at a Glance
Quick recap: scenario-based MCQ, scaled 100-1000, pass 720, 4 of 8 scenarios, no guessing penalty. Five domains, D1 the heaviest. It tests judgment — terminate on stop_reason, never parsed text.
常见问题解答
「认证概览」课时是免费的吗？
是的 — 「认证概览」的完整文本可在网页上免费阅读。要进行交互式练习（内置代码编辑器和全天候 AI 导师）并解锁 Claude Architect 课程的其余内容，请升级到 CoddyKit PRO。 Claude Architect 课程共包含 4 节课。
「认证概览」这节课中我会学到什么？
了解 Claude Certified Architect 能证明什么，以及它为何重要。 你通过在浏览器中直接运行的动手代码来练习 Claude Architect，全天候 AI 导师会在你学习这节课的过程中回答你的问题。
学习 Claude Architect 需要有经验吗？
无需任何先前经验。CoddyKit 上的 Claude Architect 课程适合初学者到高级学习者，你可以从这里开始或从头开始，按照自己的节奏学习。 这是第 1 节课，共 4 节。
「认证概览」课时需要多长时间？
大多数 CoddyKit 课程大约需要 5–10 分钟。每节课都很精短且互动，所以你能稳步进步，并在网页和应用中从离开的地方继续。
我能在这节 Claude Architect 课中编写并运行代码吗？
能。每节 Claude Architect 课都包含内置代码编辑器，你可以在浏览器中直接编写并运行真实代码，并获得即时 AI 反馈 — 无需本地设置。