gh300-lab-creator
Create short GH-300 practice exercises (10-20 minutes) that are executable and self-validating. Every exercise includes prerequisites, exact tasks, validation steps, expected outcomes, and rollback. Use when the user asks for a hands-on exercise, practice flow, or guided walkthrough.
ソース情報
- リポジトリ
- timothywarner-org/copilot-cert-prep
- ソースの最終更新活動
- 2026年9月23日 14:13
- 検出された SKILL.md の言語
- 英語
- スター
- 97
- フォーク
- 71
インストール方法
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
ソースファイルを確認
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
ファイルエクスプローラー
5 ファイルSKILL.md を表示中
SKILL.md
ソースの指示 · 読み取り専用プレビュー- name
- gh300-lab-creator
- description
- Create short GH-300 practice exercises (10-20 minutes) that are executable and self-validating. Every exercise includes prerequisites, exact tasks, validation steps, expected outcomes, and rollback. Use when the user asks for a hands-on exercise, practice flow, or guided walkthrough.
# Skill: gh300.practice_exercises.micro.validated
**Description:** Create short GH-300 practice exercises (10-20 minutes) that are executable and self-validating.
## Grounding
**Required sources:**
- Microsoft Learn (primary truth source for feature and policy behavior; access via the **Microsoft Learn MCP server**)
- Microsoft Learn code samples for syntax when applicable
## Guardrails
- Keep the exercise within GH-300 scope.
- Prefer low-risk actions that can run in a local repo or demo org.
- No contractions.
- Always include rollback or cleanup steps.
- Always use current terminology.
## Bundled assets
Use the skill-local bundle to keep outputs consistent and auditable.
- `resources/source-pack.md`
- `resources/microsoft-voice-principles.md`
- `resources/lab-quality-checklist.md`
- `scripts/validate-output.js`
## Fictional company randomization (non-negotiable)
Use fictional company names from `references/fictional-companies.md` for any scenario context. Randomize the company selection.
## Timebox guidance
An exercise should contain no more than 12 steps total across all tasks. If content exceeds that, split into two exercises.
## Workflow
1. Choose a single GH-300 objective from `references/gh300-objectives.md` and state it at the top.
2. Ground the intended behavior in Microsoft Learn.
3. Draft steps using one primary surface (VS Code, GitHub.com, or CLI).
4. Verify command syntax if commands are included.
5. Add validation gates after each major step.
6. Add rollback that exactly reverses the changes.
7. Run the checks in `resources/lab-quality-checklist.md` and apply `scripts/validate-output.js` rules before final delivery.
## Output format
```yaml
exercise:
title: "<Action + focus area>"
objective: "<One sentence outcome tied to GH-300>"
skill_area: "<GH-300 skill area>"
estimated_time: "<10-20 min>"
prerequisites:
- "<Workspace and account requirements>"
starting_state:
- "<What must already exist>"
tasks:
- name: "<Task 1 name>"
steps: |
<Numbered steps only when sequencing matters.>
validation:
- "<Validation action + expected result>"
- name: "<Task 2 name>"
steps: |
<...>
validation:
- "<...>"
troubleshooting:
- symptom: "<common failure>"
fix: "<precise fix>"
cleanup:
steps: |
<Exact rollback steps>
validation:
- "<Check that rollback succeeded>"
references:
- "<Microsoft Learn URL(s)>"
```
## Delivery rules
Exercises are delivered in full in a single message. If multiple exercises are requested, deliver each exercise sequentially in the same message.
GitHubで見る