Study guides / CCAO-F
Build exercises
A hands-on exercise for every lesson - one build task, with hints, to prove you can actually apply the concept, not just recognise it.
Domain 1 - Prompting and Task Execution
-
Give a Claude Project a Consistent Role
From 1.1 — System Prompts and Giving Claude a Role
-
Fix Inconsistent Output with Worked Examples
From 1.2 — Multishot Prompting for Consistent Output
-
Diagnose and Fix a Tag-Confusion Prompt
From 1.3 — Structuring Prompts with XML Tags
-
Prompt Differently for a Brainstorm vs. an Analysis
From 1.4 — Adapting Prompts to Audience and Task Type
-
Decompose a Multi-Part Request
From 1.5 — Task Decomposition for Complex Requests
Domain 2 - Output Evaluation and Validation
-
Build a Red-Flag Checklist by Testing It Live
From 2.1 — Detecting Hallucinations, Inconsistencies, and Bias
-
Verify Claude's Quotes Against Their Actual Source
From 2.2 — Fact-Checking and Citation-Traceable Answers
-
Compare a Self-Graded Pass Against an Independent One
From 2.3 — The Risk of a Model Grading Its Own Output
-
Catch an Assumption Error by Reading the Reasoning Trace
From 2.4 — Reviewing Extended Thinking as a Validation Step
-
Draft a Human-Oversight Recommendation for a High-Stakes Scenario
From 2.5 — Human Oversight for High-Stakes Decisions
-
Run a Paraphrase Consistency Check
From 2.6 — Checking Semantic Consistency Across Paraphrases
-
Draft, Compare, and Format for Two Different Audiences
From 2.7 — Editing, Refining, and Comparing Outputs
Domain 3 - Product and Model Selection
-
Match Model Tiers to Task Shape
From 3.1 — Choosing a Model for the Task
-
Compare Effort Levels on the Same Task
From 3.2 — Tuning Thoroughness with the Effort Parameter
-
Practice the Feature Decision Framework
From 3.3 — Choosing the Right Claude Product Feature
-
Practice Restarting, Summarising, and Persisting
From 3.4 — Context Limits: When to Restart, Summarise, or Persist
Domain 4 - Workflow Integration and Solution Design
-
Sort a Task List by Fit
From 4.1 — Analyzing Requirements Before Reaching for Claude
-
Scope a Research Task and Turn It Into a Plan
From 4.2 — Using Claude for Research and Planning
-
Map Augment vs. Redesign for a Real Process
From 4.3 — Process Optimisation: Augment vs. Redesign
-
Prototype and Iterate a Recurring-Task Project
From 4.4 — Designing and Iterating on a Claude-Based Solution
-
Design the Handoffs Around a Claude-Drafted Output
From 4.5 — Integrating Claude into an Existing Workflow
-
Write a Stakeholder Summary That Avoids Both Traps
From 4.6 — Communicating Claude's Value and Limitations to Stakeholders
Domain 5 - Configuration and Knowledge Management
-
Configure a Project's Instructions and Knowledge Base
From 5.1 — Configuring a Claude Project: Instructions and Knowledge Sources
-
Use a Connector to Read Live Content and Write Back to Drive
From 5.2 — Connectors: Drive, Gmail, and Other Data Sources
-
Generalize a Task-Specific Instruction Set
From 5.3 — Writing Effective Project-Level Instructions
-
Audit and Refresh a Project's Configuration
From 5.4 — Maintaining Configurations and Knowledge Over Time
Domain 6 - Governance, Risk, and Responsible Use
-
Evaluate a Business Use Case Against the Human-in-the-Loop Test
From 6.1 — Appropriate and Inappropriate Use Cases
-
Match Console Roles to Job Needs and Justify with Governance Reasoning
From 6.2 — Console Roles and Scoped Permissions
-
Weigh Convenience vs. Control in Project Sharing
From 6.3 — Project Sharing and Access Governance
-
Distinguish Inference Location, Storage Location, and Data Sensitivity
From 6.4 — Data Residency and Privacy Considerations
-
Locate the Activity Feed and the Transparency Hub, and Apply Them to an Incident Scenario
From 6.5 — Audit Trails and the Transparency Hub
Domain 7 - Troubleshooting and Optimisation
-
Run a Diagnosis Loop on a Real Prompt
From 7.1 — Diagnosing an Underperforming Prompt
-
Compare Two Prompt Versions Against a Fixed Test Set
From 7.2 — Adjusting Approach Based on Feedback and Results
-
Run the Diagnostic Checklist on a Real Bad Output
From 7.3 — A Decision Framework for Poor Output
-
Turn a Recurring Task into a Reusable, Right-Sized Workflow
From 7.4 — Optimising a Workflow for Efficiency