Hands-on Lab — CCA-F Claude Certified Architect — Foundations

Overview

Prerequisites

Steps

1. 1.Install Claude Code and confirm it can reach the API

   Provisions:

   • Claude Code
   • Anthropic API

   Claude Code is the CLI that the CCA-F exam tests most directly — configuration, hooks, slash commands, and permission modes all live here. We start by installing it globally so every later step can call claude.

   After installation, a quick claude --version proves the binary is on your PATH, and claude --print-system-prompt confirms the API key is valid (it makes a lightweight call under the hood).

   Copy

   # Install Claude Code globally
   npm install -g @anthropic-ai/claude-code
   
   # Verify installation
   claude --version
   
   # Confirm API connectivity (prints the built-in system prompt)
   claude --print-system-prompt | head -20

2. 2.Set up the CLAUDE.md project-level and user-level files

   Provisions:

   • CLAUDE.md
   • Project Config

   The exam heavily tests the three-tier CLAUDE.md hierarchy: user (~/.claude/CLAUDE.md), project (repo root), and module (subdirectories). Each level inherits from the one above and can override it.

   We will create a project-level file that sets coding conventions and a user-level file that sets personal preferences. When Claude Code reads the project, it merges all three tiers — exactly the behavior the exam asks about.

   Copy

   # Create project root CLAUDE.md
   cat > CLAUDE.md << 'EOF'
   # Project: CCA-F Lab
   
   ## Coding conventions
   - Use TypeScript for all new files
   - Prefer const over let
   - Use single quotes for strings
   - No semicolons (Prettier default)
   
   ## Testing
   - Run `npm test` before committing
   - All new functions must have at least one test
   EOF
   
   # Create user-level CLAUDE.md (applies to ALL your projects)
   mkdir -p ~/.claude
   cat > ~/.claude/CLAUDE.md << 'EOF'
   # User preferences
   - Be concise in responses
   - Prefer functional programming patterns
   - Always explain the "why" before showing code
   EOF
   
   # Verify Claude Code sees both files
   claude "What instructions do you see in CLAUDE.md?" --print

3. 3.Build a custom slash command that scaffolds a new module

   Provisions:

   • Slash Commands
   • Claude Code

   Custom slash commands live in .claude/commands/ as Markdown files. The filename becomes the command name (e.g. scaffold.md → /scaffold). The file content is the prompt template — it can include $ARGUMENTS for user input.

   This is a high-frequency exam topic: where commands live, how arguments are passed, and how they differ from hooks.

   Copy

   # Create the commands directory
   mkdir -p .claude/commands
   
   # Create a scaffold command
   cat > .claude/commands/scaffold.md << 'EOF'
   Create a new TypeScript module named $ARGUMENTS with:
   1. A main source file at src/$ARGUMENTS/index.ts
   2. A test file at src/$ARGUMENTS/__tests__/index.test.ts
   3. Export the module from the project root index.ts
   
   Follow the coding conventions in CLAUDE.md.
   EOF
   
   # Test the command (Claude Code will execute it interactively)
   claude /scaffold calculator

4. 4.Configure hooks to auto-lint before every file write

   Provisions:

   • Hooks
   • Settings

   Hooks run shell commands in response to Claude Code events. The exam tests three hook types: PreToolUse (before a tool runs), PostToolUse (after), and Notification (on status changes). Hooks are defined in .claude/settings.json.

   We will add a PostToolUse hook that runs a linter after every file edit — a common production pattern that the exam frequently asks about.

   Copy

   # Initialize a package.json so we have a lint script
   npm init -y
   npm install --save-dev typescript @typescript-eslint/parser
   
   # Create project-level settings with a post-tool hook
   mkdir -p .claude
   cat > .claude/settings.json << 'EOF'
   {
     "hooks": {
       "PostToolUse": [
         {
           "matcher": "Write|Edit",
           "command": "npx tsc --noEmit 2>&1 | head -5"
         }
       ]
     },
     "permissions": {
       "allow": [
         "Read",
         "Write",
         "Edit"
       ]
     }
   }
   EOF
   
   # Verify settings are recognized
   cat .claude/settings.json

5. 5.Spin up a local MCP server and connect Claude Code to it

   Provisions:

   • MCP
   • stdio Transport

   The Model Context Protocol (MCP) is a core exam domain. MCP servers expose tools, resources, and prompts to LLM clients via a standardized JSON-RPC protocol. The stdio transport is the simplest — the server reads from stdin and writes to stdout.

   We will build a minimal MCP server that exposes a get_weather tool, then configure Claude Code to connect to it. This exercises the full MCP lifecycle the exam tests: tool definition, transport wiring, tool invocation, and result handling.

   Copy

   # Install the MCP SDK
   npm install @modelcontextprotocol/sdk
   
   # Create the MCP server
   cat > mcp-weather.mjs << 'EOF'
   import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js";
   import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js";
   
   const server = new McpServer({ name: "weather", version: "1.0.0" });
   
   server.tool("get_weather", { city: { type: "string" } }, async ({ city }) => ({
     content: [{ type: "text", text: `Weather in ${city}: 72°F, sunny` }],
   }));
   
   const transport = new StdioServerTransport();
   await server.connect(transport);
   EOF
   
   # Register the MCP server in Claude Code settings
   cat > .claude/settings.json << 'EOF'
   {
     "mcpServers": {
       "weather": {
         "command": "node",
         "args": ["mcp-weather.mjs"]
       }
     }
   }
   EOF
   
   # Ask Claude Code to use the MCP tool
   claude "What is the weather in Tokyo? Use the get_weather tool."

6. 6.Call the Anthropic API with tool definitions and structured output

   Provisions:

   • Tool Use API
   • Structured Output

   The exam tests the Messages API tool-use flow in detail: you send tools in the request, the model responds with a tool_use content block, you execute it and return a tool_result, then the model produces the final answer.

   This step uses curl to call the API directly — the most exam-aligned approach since the exam tests the raw request/response format, not SDK wrappers.

   Copy

   # Call the Messages API with a tool definition
   curl -s https://api.anthropic.com/v1/messages \
     -H "x-api-key: $ANTHROPIC_API_KEY" \
     -H "anthropic-version: 2023-06-01" \
     -H "content-type: application/json" \
     -d '{
       "model": "claude-sonnet-4-6",
       "max_tokens": 1024,
       "tools": [{
         "name": "calculate",
         "description": "Evaluate a math expression",
         "input_schema": {
           "type": "object",
           "properties": {
             "expression": { "type": "string", "description": "Math expression to evaluate" }
           },
           "required": ["expression"]
         }
       }],
       "messages": [{
         "role": "user",
         "content": "What is 15% tip on a $85.50 dinner bill?"
       }]
     }' | python3 -m json.tool

7. 7.Enable extended thinking for complex multi-step reasoning

   Provisions:

   • Extended Thinking
   • Anthropic API

   Extended thinking lets the model use a thinking block to reason step-by-step before responding. The exam tests when to use it (complex reasoning, math, code generation) and how to configure budget_tokens.

   We send a deliberately tricky logic puzzle and compare responses with and without extended thinking to see the quality difference firsthand.

   Copy

   # Request WITH extended thinking
   curl -s https://api.anthropic.com/v1/messages \
     -H "x-api-key: $ANTHROPIC_API_KEY" \
     -H "anthropic-version: 2023-06-01" \
     -H "content-type: application/json" \
     -d '{
       "model": "claude-sonnet-4-6",
       "max_tokens": 16000,
       "thinking": {
         "type": "enabled",
         "budget_tokens": 10000
       },
       "messages": [{
         "role": "user",
         "content": "A farmer has 17 sheep. All but 9 run away. How many does he have left? Explain your reasoning step by step."
       }]
     }' | python3 -m json.tool
   
   # The response includes a "thinking" block showing the model's
   # internal reasoning before the final answer.

8. 8.Use prompt caching to cut costs on repeated system prompts

   Provisions:

   • Prompt Caching
   • Anthropic API

   Prompt caching lets you mark parts of the request as cacheable. Cached prefixes cost 90% less on subsequent calls and have zero additional latency. The exam tests the cache_control block placement and the cache-hit headers.

   We send the same large system prompt twice and check the response headers to confirm the cache hit on the second call.

   Copy

   # First call — creates the cache entry
   curl -s -D /dev/stderr https://api.anthropic.com/v1/messages \
     -H "x-api-key: $ANTHROPIC_API_KEY" \
     -H "anthropic-version: 2023-06-01" \
     -H "content-type: application/json" \
     -d '{
       "model": "claude-sonnet-4-6",
       "max_tokens": 256,
       "system": [{
         "type": "text",
         "text": "You are a helpful assistant specialized in cloud architecture. You follow AWS Well-Architected Framework principles. You always recommend infrastructure as code. You prefer Terraform over CloudFormation. You prioritize security and cost optimization.",
         "cache_control": { "type": "ephemeral" }
       }],
       "messages": [{ "role": "user", "content": "How should I set up VPC peering?" }]
     }' > /dev/null 2>&1
   
   # Second call — should hit the cache (check usage.cache_read_input_tokens)
   curl -s https://api.anthropic.com/v1/messages \
     -H "x-api-key: $ANTHROPIC_API_KEY" \
     -H "anthropic-version: 2023-06-01" \
     -H "content-type: application/json" \
     -d '{
       "model": "claude-sonnet-4-6",
       "max_tokens": 256,
       "system": [{
         "type": "text",
         "text": "You are a helpful assistant specialized in cloud architecture. You follow AWS Well-Architected Framework principles. You always recommend infrastructure as code. You prefer Terraform over CloudFormation. You prioritize security and cost optimization.",
         "cache_control": { "type": "ephemeral" }
       }],
       "messages": [{ "role": "user", "content": "What about NAT gateway sizing?" }]
     }' | python3 -c "import sys,json; d=json.load(sys.stdin); print(f\"Cache read tokens: {d['usage'].get('cache_read_input_tokens', 0)}\")"

9. 9.Run Claude Code in headless mode for CI/CD pipelines

   Provisions:

   • Headless Mode
   • CI/CD

   The exam tests how to run Claude Code non-interactively — the --print flag outputs the result to stdout without an interactive session, and --output-format json gives structured output for scripting.

   This is the pattern used in CI pipelines: feed a prompt, get a result, parse it programmatically.

   Copy

   # Headless mode: prompt in, text out
   claude --print "List 3 TypeScript best practices, one per line"
   
   # JSON output for programmatic use
   claude --print --output-format json "What is 2+2?"
   
   # Pipe a file for code review (CI pattern)
   echo 'const x: any = "hello";' > example.ts
   claude --print "Review this TypeScript file for type safety issues: $(cat example.ts)"

10. 10.Build a multi-agent supervisor that delegates to specialist sub-agents

    Provisions:

    • Multi-Agent
    • Supervisor Pattern
    • Anthropic SDK

    Domain 1 (Agentic Architecture, 27%) is the highest-weighted on the exam. The supervisor pattern — one orchestrator dispatching to specialist sub-agents, each with its own system prompt and tool set — is the most-tested architecture.

    We will build a self-contained Node.js script that implements a supervisor coordinating two specialists: a researcher that reads files and summarizes, and a writer that produces output. The supervisor decomposes a task, delegates to each specialist via separate API calls, and aggregates the results. Run it from the terminal and watch the delegation happen live.

    Copy

    # Install the Anthropic SDK
    npm install @anthropic-ai/sdk
    
    # Create sample data for the agents to work with
    cat > sales-data.csv << 'EOF'
    month,revenue,units
    Jan,12500,150
    Feb,15200,180
    Mar,11800,140
    Apr,18900,220
    May,22100,260
    Jun,19500,230
    EOF
    
    # Create the multi-agent supervisor script
    cat > supervisor.mjs << 'SCRIPT'
    import Anthropic from "@anthropic-ai/sdk";
    import { readFileSync } from "fs";
    
    const client = new Anthropic();
    
    async function callAgent(role, systemPrompt, userMessage) {
      console.log(`\n--- ${role} agent ---`);
      const response = await client.messages.create({
        model: "claude-sonnet-4-6",
        max_tokens: 1024,
        system: systemPrompt,
        messages: [{ role: "user", content: userMessage }],
      });
      const text = response.content
        .filter((b) => b.type === "text")
        .map((b) => b.text)
        .join("\n");
      console.log(text);
      return text;
    }
    
    // --- Supervisor: decompose, delegate, aggregate ---
    const data = readFileSync("sales-data.csv", "utf-8");
    
    console.log("=== SUPERVISOR: Decomposing task ===");
    console.log("Task: Analyze sales data and write a brief executive summary.\n");
    console.log("Step 1: Delegate analysis to Researcher agent");
    console.log("Step 2: Delegate writing to Writer agent");
    console.log("Step 3: Combine results\n");
    
    // Sub-agent 1: Researcher — extracts insights
    const analysis = await callAgent(
      "Researcher",
      "You are a data analyst. Extract key trends, highs, lows, and growth rates. Be precise with numbers. Output bullet points only.",
      `Analyze this CSV data:\n\n${data}`
    );
    
    // Sub-agent 2: Writer — produces the summary from the analysis
    const summary = await callAgent(
      "Writer",
      "You are a business writer. Write a 3-sentence executive summary from the analysis provided. Use a professional tone. Include specific numbers.",
      `Write an executive summary based on these findings:\n\n${analysis}`
    );
    
    console.log("\n=== SUPERVISOR: Final aggregated output ===");
    console.log(summary);
    SCRIPT
    
    # Run the supervisor
    node supervisor.mjs

Cleanup

# Remove the lab directory
cd .. && rm -rf cca-f-lab

# Optionally remove the user-level CLAUDE.md we created
rm -f ~/.claude/CLAUDE.md

# Optionally uninstall Claude Code
npm uninstall -g @anthropic-ai/claude-code