Guided labs for the CCA Foundations certification exam.
Build working code for each of the 6 exam scenarios so you can apply the concepts before test day, not just study them. The labs cover all 5 domains and all 30 task statements from the official exam guide.
| Lab | Description | Coverage |
|---|---|---|
| 01 Customer Support Resolution Agent |
A customer support agent with four tools, a manual agentic loop, a prerequisite gate, and a PostToolUse hook enforcing refund policy, using the Claude API. | D1, D2, D5 1.1, 1.2, 1.4, 1.5, 2.1, 2.2, 2.3, 5.1, 5.2 |
| 02 Code Generation with Claude Code |
A Claude Code workspace with CLAUDE.md hierarchy, path-specific rules, a context:fork skill, MCP servers, and session management with resume and fork. |
D3, D5 1.7, 2.4, 3.1, 3.2, 3.3, 3.4, 3.5, 5.4 |
| 03 Multi-Agent Research System |
A coordinator that delegates to four research subagents, handles structured errors from unavailable sources, and synthesizes cited reports with coverage gap annotations, using the Claude Agent SDK. | D1, D2, D5 1.2, 1.3, 1.5, 1.6, 2.1, 2.2, 2.3, 5.1, 5.3, 5.6 |
| 04 Developer Productivity with Claude |
A codebase exploration agent with built-in tools, an MCP docs server, an Explore subagent, scratchpad persistence, and a context:fork skill, using the Claude Agent SDK. |
D1, D2, D3, D5 1.3, 2.1, 2.4, 2.5, 3.1, 3.2, 3.4, 5.4 |
| 05 Claude Code for Continuous Integration |
A simulated CI pipeline using Claude Code's -p flag: iterative prompt refinement, per-file and cross-file review passes, independent review instances, and CLAUDE.md project context. |
D3, D4 1.6, 3.4, 3.5, 3.6, 4.1, 4.2, 4.6 |
| 06 Structured Data Extraction |
An invoice extraction pipeline with forced tool_choice, nullable fields, self-correction, validation-retry loops, few-shot examples, batch processing, and confidence-based human review routing, using the Claude API. |
D4, D5 4.2, 4.3, 4.4, 4.5, 5.5 |
- Python 3.11+
- An Anthropic API key
- Claude Code (required for Labs 02 and 05)
cd 0X_lab_name
python -m venv .venv
source .venv/bin/activate # macOS/Linux
.venv\Scripts\activate # Windows
cp .env.example .env # add your ANTHROPIC_API_KEY
pip install -r requirements.txt
python main.py # run the lab
These labs were carefully designed to cover every domain and task from the official exam guide, with a guided, educational approach that walks you through each concept step by step.
See the Lab Reference for details of what domains and tasks are covered by each lab.
| ID | Domain | Weight | Labs |
|---|---|---|---|
| D1 | Agentic Architecture & Orchestration | 27% | 01, 02, 03, 04, 05 |
| D2 | Tool Design & MCP Integration | 18% | 01, 02, 03, 04 |
| D3 | Claude Code Configuration & Workflows | 20% | 02, 04, 05 |
| D4 | Prompt Engineering & Structured Output | 20% | 05, 06 |
| D5 | Context Management & Reliability | 15% | 01, 02, 03, 04, 06 |
This is an independent resource, not produced, endorsed, or affiliated with Anthropic.
