File: 'analytics-team-capacity.csv'\nRows: 32 · Columns: week, team_member, hours_logged, active_projects\n\nFull data:\nWeek 1, Maria, 46, 2\nWeek 1, Josh, 44, 2\nWeek 1, Priya, 45, 3\nWeek 1, Tom, 47, 2\nWeek 2, Maria, 48, 2\nWeek 2, Josh, 47, 2\nWeek 2, Priya, 46, 3\nWeek 2, Tom, 50, 3\nWeek 3, Maria, 50, 3\nWeek 3, Josh, 49, 2\nWeek 3, Priya, 48, 3\nWeek 3, Tom, 53, 3\nWeek 4, Maria, 52, 3\nWeek 4, Josh, 51, 3\nWeek 4, Priya, 50, 3\nWeek 4, Tom, 56, 3\nWeek 5, Maria, 54, 3\nWeek 5, Josh, 53, 3\nWeek 5, Priya, 52, 3\nWeek 5, Tom, 59, 3\nWeek 6, Maria, 55, 3\nWeek 6, Josh, 54, 3\nWeek 6, Priya, 53, 3\nWeek 6, Tom, 61, 4\nWeek 7, Maria, 56, 3\nWeek 7, Josh, 55, 3\nWeek 7, Priya, 54, 3\nWeek 7, Tom, 62, 4\nWeek 8, Maria, 57, 3\nWeek 8, Josh, 56, 3\nWeek 8, Priya, 55, 3\nWeek 8, Tom, 63, 4\n\nAuto-detected: team average climbed from 45.5 hrs/week in Week 1 to 57.75 hrs/week in Week 8, with no week showing a decline. Tom has exceeded 60 hours for four straight weeks and is now carrying 4 active projects, the highest sustained load on the team.

Summary  
This chapter covers detecting and interpreting trends in real-number time-series data by calculating metrics like capacity variation to assess whether changes are temporary or sustained.

General domain of usage  
Project resource management



The prompt-only version guessed that the team is overloaded because you said so in one sentence. Whether that's actually a sustained trend or just a rough couple of weeks is exactly the kind of thing you shouldn't be guessing at when you're about to ask for headcount.



> ## Task
> You'll work with the capacity data below. Ask Claude to actually interpret it:
>
> > <b>Analyse our team's capacity data. Is the workload increase temporary or a sustained trend? What's the risk if we don't hire?

**Analytics Team Capacity Log**

| Week | Team Member | Hours Logged | Active Projects |
|---|---|---|---|
| Week 1 | Maria | 46 | 2 |
| Week 1 | Josh | 44 | 2 |
| Week 1 | Priya | 45 | 3 |
| Week 1 | Tom | 47 | 2 |
| Week 2 | Maria | 48 | 2 |
| Week 2 | Josh | 47 | 2 |
| Week 2 | Priya | 46 | 3 |
| Week 2 | Tom | 50 | 3 |
| Week 3 | Maria | 50 | 3 |
| Week 3 | Josh | 49 | 2 |
| Week 3 | Priya | 48 | 3 |
| Week 3 | Tom | 53 | 3 |
| Week 4 | Maria | 52 | 3 |
| Week 4 | Josh | 51 | 3 |
| Week 4 | Priya | 50 | 3 |
| Week 4 | Tom | 56 | 3 |
| Week 5 | Maria | 54 | 3 |
| Week 5 | Josh | 53 | 3 |
| Week 5 | Priya | 52 | 3 |
| Week 5 | Tom | 59 | 3 |
| Week 6 | Maria | 55 | 3 |
| Week 6 | Josh | 54 | 3 |
| Week 6 | Priya | 53 | 3 |
| Week 6 | Tom | 61 | 4 |
| Week 7 | Maria | 56 | 3 |
| Week 7 | Josh | 55 | 3 |
| Week 7 | Priya | 54 | 3 |
| Week 7 | Tom | 62 | 4 |
| Week 8 | Maria | 57 | 3 |
| Week 8 | Josh | 56 | 3 |
| Week 8 | Priya | 55 | 3 |
| Week 8 | Tom | 63 | 4 |

Spreadsheet for the Task

Master the skills needed to build reliable AI decision systems with Claude, tackling real business calls like switching tools, approving a hire, and renewing a vendor contract. Progress from prompt-only reasoning to tool-grounded analysis with live data, then package your approach into a reusable Claude agent with structured output and a no-invented-facts constraint. By the end, you'll write a full agent system prompt from scratch.

Discover how to build an AI decision system for a real tool-switch call. Start with prompt-only reasoning, ground it in live pricing and migration data via web search, then build a Tool Switch Advisor agent that routes tasks and cites its sources.

Explore the essentials of contextual prompting for hiring decisions. Compare a bare prompt to a context-rich one, analyze real team capacity data to test the case for a hire, then build a Hiring Decision Advisor agent around that data.

Dive into the fundamentals of building a Claude agent for a vendor renewal decision. Combine past notes, usage data, and market research into one recommendation, then write your own agent prompt with tool routing and a no-invented-facts rule.