← Back to list

aidlc-workflows
by aws-samples
A collection of AWS Korea Startup SA team materials for hands-on labs.
⭐ 46🍴 26📅 Jan 23, 2026
SKILL.md
name: aidlc-workflows description: AI-DLC workflow execution that loads and follows external aidlc-workflows rules to plan, design, and generate code with approval gates and aidlc-docs artifacts. Use when a user wants to run AI-DLC, needs adaptive inception/construction phases, or needs rule-driven requirements, design, and code generation from aidlc-workflows.
AIDLC Workflow Skill
Core behavior
- Locate the
aidlc-workflows/repository and read rules from it on demand. - Follow AI-DLC phases (inception, construction, operations) with explicit approval gates.
- Write artifacts to
aidlc-docs/and preserve audit/state across phases. - Never modify rule files under
aidlc-workflows/aidlc-rules/.
Locate aidlc-workflows
Use this lookup order:
AIDLC_WORKFLOWS_DIRenvironment variable, if set.aidlc-workflows/in the current workspace.~/aidlc-workflows.
If none are found, ask the user in English using this exact prompt:
"I couldn't find the aidlc-workflows directory. Please provide the absolute path to your aidlc-workflows repository (for example: /Users/yourname/path/to/aidlc-workflows)."
Rule loading
- Read
aidlc-workflows/skills/ai-dlc/SKILL.mdfor baseline workflow intent. - Use
aidlc-workflows/skills/ai-dlc/loader.mdas the source of truth for dynamic rule loading. - Load only the specific rule files required for the current stage.
- Prefer common rules first, then phase-specific rules, then stage-specific rules.
See references/loader.md for the loading sequence summary and key rule files.
Execution flow
- Confirm workspace state (greenfield vs brownfield) and initialize
aidlc-docs/. - Run INCEPTION stages per rules (always/conditional) and request approval after each stage.
- Run CONSTRUCTION per unit, including design, code generation, and build/test stages.
- If OPERATIONS is requested, follow the operations rule file when present.
- Keep a clear audit trail in
aidlc-docs/audit.mdand state inaidlc-docs/aidlc-state.md.
Output rules
- Use the question/answer format required by the rule files when asking clarifying questions.
- If a stage is skipped, document the skip reason in the audit trail.
- Never copy rule content into the skill; only reference and load it as needed.
Example trigger phrases
- "Using AI-DLC, build a user authentication service with OAuth 2.0"
- "/aidlc-workflows Add a password reset feature"
Score
Total Score
70/100
Based on repository quality metrics
✓SKILL.md
SKILL.mdファイルが含まれている
+20
✓LICENSE
ライセンスが設定されている
+10
○説明文
100文字以上の説明がある
0/10
○人気
GitHub Stars 100以上
0/15
✓最近の活動
1ヶ月以内に更新
+10
✓フォーク
10回以上フォークされている
+5
✓Issue管理
オープンIssueが50未満
+5
✓言語
プログラミング言語が設定されている
+5
✓タグ
1つ以上のタグが設定されている
+5
Reviews
💬
Reviews coming soon
