# AB-100 Exam Drill — Scout Skill

An interactive daily drill for **AB-100: Agentic AI Business Solutions Architect**.

Scout serves you 20 blueprint-weighted, scenario-based questions one at a time, grounded in live Microsoft Learn docs. After each answer it tells you why the right answer is right, why each distractor is wrong, and where to read more. It tracks accuracy per objective in a local SQLite file and biases future sessions toward your weak areas.

Built and used to pass the real exam. Everything runs locally; nothing leaves your machine.

## Install

1. Unzip so the `ab-100` folder sits inside your Scout skills directory:

   - **Windows:** `%USERPROFILE%\.scout\m-skills\ab-100\`
   - **macOS/Linux:** `~/.scout/m-skills/ab-100/`

   You should end up with `.../m-skills/ab-100/SKILL.md`, `drill.py`, `blueprint.json`.

2. Restart Scout (or start a new session) so the skill is picked up.

3. Type `/ab-100` in Scout.

Requires Python 3.8+ on PATH. No pip packages needed — standard library only.

## First run

On your first `/ab-100`, Scout asks for your **target exam date**, then writes `config.json` beside the script and creates `progress.db`. Everything after that is just drilling.

If your exam moves, say "my exam date changed" and Scout re-runs setup.

## Daily rhythm

One 20-question session a day is the intended cadence. Under a week out, two.

Useful things to say:

| Say | Result |
|---|---|
| `/ab-100` or "quiz me" | Full 20-question session |
| "just 10 questions" | Shorter session |
| "readiness" / "how am I doing" | Blueprint-weighted readiness report |
| "history" | Score trend across sessions |
| "my exam date changed" | Reschedule |
| "start over" | Wipe progress and reconfigure |

Mid-question you can push back — "B, but why not C?" or "explain the difference first". Scout holds on that question until the thread is closed, then resumes.

## What's in the box

| File | Purpose |
|---|---|
| `SKILL.md` | The skill definition Scout reads |
| `drill.py` | Session planner, answer recorder, readiness reporter |
| `blueprint.json` | All 74 exam objectives with official domain weights (skills-measured dated 2026-07-22) |
| `config.json` | Created on first run — your exam date |
| `progress.db` | Created on first run — your answer history |

`config.json` and `progress.db` are yours alone. Don't share them; do back them up if you care about the trend.

## Manual CLI (optional)

```
cd <skill folder>
python drill.py status
python drill.py setup --exam-date 2026-11-04 --name "Your Name"
python drill.py report
python drill.py history
python drill.py reset --confirm
```

## A note on accuracy

The blueprint tracks the published skills-measured list as of **2026-07-22**. Microsoft revises these. Before your exam, check the official
[AB-100 study guide](https://learn.microsoft.com/credentials/certifications/resources/study-guides/ab-100)
and update `blueprint.json` if objectives have shifted. The readiness band is a coaching signal, not a score prediction.
