ai-developer-skill-os 3.1.5 → 5.0.0

package/skills/qk-ai-builder/SKILL.md

@@ -1,48 +1,57 @@
 ---
 name: qk-ai-builder
-version: 3.1.5
-updated: 2026-07-02
+version: 5.0.0
+updated: 2026-07-03
 description: Thiết kế AI Logic, Prompts, RAG pipelines, và các Agents.
 category: ai
-behavior: development
-intent: implement-feature
-priority: medium
 tags: [ai, llm, prompt-engineering, rag, agents]
 platforms: [claude-code, cursor, windsurf, gemini-cli]
-trigger: User asks to build a chatbot, add AI features, or write prompts.
-inputs: [AI Requirements]
-outputs: [AI Logic, Prompts, RAG setup]
-allowed_tools: [run_command, write_to_file]
-pipeline: [analyze, design, implement, validate, complete]
 ---
 
-# 🛠️ qk-ai-builder - Standard Operating Procedure
+# 🛠️ qk-ai-builder
 
-> **Goal:** "AI Engineer". Xây dựng các tính năng thông minh bằng LLM (Chatbot, RAG pipeline, Agent workflows).
+> **Inheritance:** Kỹ năng này tuân thủ Kiến trúc v5.0 của `framework/KERNEL.md`.
+> Output bắt buộc là Decision Summary.
 
-## 🔄 1. Chain of Thought (SOP)
+---
+
+## 🎯 Mission (Scope)
+- ✅ Xây dựng luồng AI/LLM, viết Prompts chuyên nghiệp.
+- ❌ Do NOT dùng LLM để thay thế các logic cứng có thể giải quyết bằng Code (như tính toán).
 
-1. **Analyze (Task Breakdown):**
-   - Determine the AI use case (Classification, Generation, Chat, Search).
-2. **Design (Prompt Engineering):**
-   - Write clear System Prompts. Treat prompts like code (use variables, constraints).
-3. **Implement (Integration):**
-   - Use the appropriate SDK (LangChain, OpenAI SDK, Vercel AI).
-4. **Validate (Safety Check):**
-   - Ensure API keys are not hardcoded.
+---
 
-## 🛡️ 2. Constraints & Rules
+## ⚙️ Capabilities (Cognitive Pipeline)
+```yaml
+Pipeline:
+  - inference
+  - planning
+  - execution
+  - bias-review
+  - ship-check
+```
 
-- **JSON Parsing:** Always enforce JSON outputs if the AI response is consumed by the app, and handle parse errors.
+---
 
-## 🤝 3. Handoff Pipeline
+## 🎛️ Dials (Hành vi)
+```yaml
+Dials:
+  - id: complexity-budget
+```
 
-1. `validate`: Send to validation gate.
-2. `complete`: Output the AI Integration Report.
+---
 
-## 📝 4. Output Format
+## 🛡️ Biases (Sửa lỗi mặc định)
+```yaml
+Biases:
+  - id: swallow-errors # (Bỏ qua lỗi API gọi LLM thất bại)
+  - id: hallucinated-tools # (Tự bịa model không tồn tại)
+```
 
-Vui lòng trả kết quả bằng Tiếng Việt.
+---
 
-- **Tóm tắt (Summary):** Luồng AI đã thiết lập.
-- **Chi tiết (Changes):** Prompt và logic gọi API.
+## 🛫 Ship Criteria
+```yaml
+Rules:
+  - id: end-to-end-validation
+```

package/skills/qk-api-lifecycle/SKILL.md

@@ -1,72 +1,57 @@
 ---
 name: qk-api-lifecycle
-version: 3.1.5
-updated: 2026-07-02
+version: 5.0.0
+updated: 2026-07-03
 description: Thiết kế, triển khai, và tích hợp các API endpoints.
-category: engineering
-behavior: development
-intent: implement-feature
-priority: high
-tags: [api, backend, frontend-integration, endpoints]
+category: backend
+tags: [api, rest, graphql, integration]
 platforms: [claude-code, cursor, windsurf, gemini-cli]
-trigger: User wants to create a new API, fix an API, or integrate frontend with backend API.
-inputs: [API requirements, JSON schema, or Backend URL]
-outputs: [API routes, Service classes, API Hooks]
-allowed_tools: [run_command, read_file, write_to_file, grep_search]
-pipeline: [analyze, implement, engineering-standard, validate, complete]
 ---
 
-# 🛠️ qk-api-lifecycle - Standard Operating Procedure
+# 🛠️ qk-api-lifecycle
 
-> **Goal:** Quản lý toàn bộ vòng đời của một API, từ việc thiết kế Spec (Swagger/Types) cho đến khi Backend code xong và Frontend gọi thành công.
+> **Inheritance:** Kỹ năng này tuân thủ Kiến trúc v5.0 của `framework/KERNEL.md`.
+> Output bắt buộc là Decision Summary.
 
-## 🔄 1. Chain of Thought (SOP)
+---
 
-1. **Analyze (Contract Definition):**
-   - Read the user requirements.
-   - Define the exact API contract (Request Body, Query Params, Response DTO) in TypeScript interfaces or OpenAPI spec.
-2. **Plan (Architecture Selection):**
-   - Determine if the task is Backend (creating the endpoint) or Frontend (consuming the endpoint).
-   - If Frontend: Load `knowledge/frontend/react.md` to decide between React Query vs raw fetch.
-   - If Backend: Choose the appropriate layer (Controller -> Service -> Repository).
-3. **Execute (Implementation):**
-   - Write the backend handler/controller with strict validation (e.g., Zod, Class Validator).
-   - **Security:** Integrate Auth Middleware (JWT/RBAC) before the Controller. DO NOT expose public endpoints unless explicitly requested.
-   - Write the frontend API client (e.g., Axios instance wrapper).
-4. **Verify (Validation):**
-   - Handle Error Status Codes (400, 401, 403, 404, 500) gracefully.
-   - Ensure loading states are managed on the frontend.
+## 🎯 Mission (Scope)
+- ✅ Thiết kế RESTful hoặc GraphQL API chuẩn mực.
+- ❌ Do NOT trả về mã `200 OK` cho các response bị lỗi `error: true`.
 
-## 🛡️ 2. Constraints & Rules
+---
 
-- **No Hardcoded URLs:** Never hardcode `http://localhost:3000` inside frontend components. Always use environment variables (`process.env.API_URL`).
-- **Separation of Concerns:** Do NOT write `fetch` or `axios.get` directly inside a React UI component. Extract it to a custom hook or an API service file.
-- **Pagination & Rate Limit:** Any `GET` API returning a list MUST include Pagination by default to prevent server OOM.
+## ⚙️ Capabilities (Cognitive Pipeline)
+```yaml
+Pipeline:
+  - inference
+  - planning
+  - execution
+  - bias-review
+  - ship-check
+```
 
-## 🌳 3. Decision Tree
+---
 
-```text
-Is this a Backend or Frontend task?
-  ├── BACKEND → Does the API need Database access?
-  │       ├── YES → Handoff to `qk-data-lifecycle` to create models first.
-  │       └── NO → Write Controller and Service.
-  └── FRONTEND → Are we fetching data to display or mutating data?
-          ├── FETCH → Use React Query `useQuery` or equivalent.
-          └── MUTATE → Use React Query `useMutation` or equivalent.
+## 🎛️ Dials (Hành vi)
+```yaml
+Dials:
+  - id: rest-strictness
 ```
 
-## 🤝 4. Handoff Pipeline
-
-1. `engineering-standard`: Verify API naming conventions (e.g., `GET /users` instead of `POST /getUsers`).
-2. `validate`: Run API tests and Type-checks.
-3. `complete`: Generate the final report.
+---
 
-## 📝 5. Output Format
+## 🛡️ Biases (Sửa lỗi mặc định)
+```yaml
+Biases:
+  - id: swallow-errors
+  - id: enterprise-crud # (Dùng gRPC/Kafka cho API quá nhỏ)
+```
 
-Vui lòng trả kết quả bằng Tiếng Việt. Sử dụng template chuẩn của hệ thống:
+---
 
-- **Tóm tắt (Summary):** Các endpoint đã tạo hoặc tích hợp.
-- **Chi tiết (Changes):** File nào chứa logic API.
-- **Kiến trúc (Reasoning):** Lý do chọn thư viện hoặc cách bắt lỗi (Error handling).
-- **Xác thực (Verification):** Cách gọi thử API.
-- **Rủi ro (Risks):** Vấn đề bảo mật CORS hoặc Rate Limit (nếu có).
+## 🛫 Ship Criteria
+```yaml
+Rules:
+  - id: minimal-diff
+```

package/skills/qk-bug-resolution/SKILL.md

@@ -1,76 +1,57 @@
 ---
 name: qk-bug-resolution
-version: 3.1.5
-updated: 2026-07-02
+version: 5.0.0
+updated: 2026-07-03
 description: Xử lý và sửa các lỗi (bugs), đồng thời ngăn ngừa lỗi hồi quy.
-category: engineering
-behavior: development
-intent: fix-bug
-priority: high
-tags: [bug, debug, regression]
+category: maintenance
+tags: [bugfix, debugging, troubleshooting]
 platforms: [claude-code, cursor, windsurf, gemini-cli]
-trigger: User reports an error or crash.
-inputs: [Error log, Bug description, File path]
-outputs: [Fixed code, Verification report]
-allowed_tools: [run_command, read_file, grep_search]
-pipeline: [analyze, implement, engineering-standard, validate, complete]
 ---
 
-# 🛠️ qk-bug-resolution - Standard Operating Procedure
+# 🛠️ qk-bug-resolution
 
-> **Goal:** Định vị nguyên nhân, sửa lỗi triệt để, và đảm bảo lỗi không bị lặp lại (Regression Prevention).
+> **Inheritance:** Kỹ năng này tuân thủ Kiến trúc v5.0 của `framework/KERNEL.md`.
+> Output bắt buộc là Decision Summary.
 
-## 🔄 1. Chain of Thought (SOP)
-
-1. **Analyze (Root Cause Analysis & Reproduce):**
-   - Read the provided error log or bug description.
-   - Identify how to reproduce the bug. Ask the user for more info if it is unclear.
-   - Trace the error to the exact file and line of code.
-   - Identify if the bug is a syntax error, logic error, state issue, or boundary case.
-2. **Plan (Fix Strategy):**
-   - Determine the minimal change required to fix the bug.
-   - If the bug is UI-related, refer to frontend knowledge (e.g., `knowledge/frontend/react.md`).
-3. **Execute (Implement Fix):**
-   - Modify the source code exactly where needed.
-   - Do NOT rewrite unrelated code blocks.
-4. **Verify (Validation):**
-   - Provide steps to test the fix.
-   - Ensure backward compatibility.
-
-## 🛡️ 3. Constraints & Rules
-
-- **No Guesses:** Do not assume the shape of an API response. Print it or read the type definitions first.
-- **Minimal Fix:** Always prefer a 1-line fix over a 100-line refactor, unless the architecture is fundamentally broken.
-- **Clean Up:** Remove any `console.log()` or debugger statements introduced during the debugging process.
-- **Regression Test:** Add at least 1 Unit or Integration Test to prevent this bug from recurring.
+---
 
-## 🌳 3. Decision Tree
+## 🎯 Mission (Scope)
+- ✅ Tìm ra nguyên nhân gốc rễ (root cause) và sửa lỗi.
+- ❌ Do NOT viết lại (rewrite) toàn bộ file thay vì sửa đúng dòng gây lỗi (The Rewrite Trap).
 
-```text
-Is there a Stack Trace?
-  ├── YES → Use grep_search to find the exact file and line number.
-  └── NO → Ask the user for the file path or use grep_search with keywords from the bug description.
+---
 
-Is the bug related to State/Re-rendering?
-  ├── YES → Load `knowledge/frontend/react.md` to check for stale closures or missing dependencies.
-  └── NO → Proceed with standard logic debugging.
+## ⚙️ Capabilities (Cognitive Pipeline)
+```yaml
+Pipeline:
+  - inference
+  - planning
+  - execution
+  - bias-review
+  - ship-check
 ```
 
-## 🤝 4. Handoff Pipeline
+---
 
-After implementing the fix, map to the pipeline:
+## 🎛️ Dials (Hành vi)
+```yaml
+Dials:
+  - id: strictness
+```
 
-1. `engineering-standard`: Ensure the fix follows naming and architecture rules.
-2. `validate`: Run linters and tests (via `qk-validation-gate` equivalent).
-3. `complete`: Output the final report.
+---
 
-## 📝 5. Output Format
+## 🛡️ Biases (Sửa lỗi mặc định)
+```yaml
+Biases:
+  - id: swallow-errors
+  - id: cosmetic-refactor # (Tuyệt đối cấm refactor khi đang fix bug)
+```
 
-Vui lòng sử dụng template `templates/bug-report.md` (Báo Cáo Xử Lý Lỗi) để phản hồi lại cho người dùng bằng Tiếng Việt. Nội dung bao gồm:
+---
 
-- Summary
-- Changes
-- Root Cause
-- Verification
-- Risks
-- Next Action
+## 🛫 Ship Criteria
+```yaml
+Rules:
+  - id: minimal-diff # (Chỉ sửa dòng cần sửa)
+```

package/skills/qk-context-loader/SKILL.md

@@ -1,55 +1,56 @@
 ---
 name: qk-context-loader
-version: 3.1.5
-updated: 2026-07-02
+version: 5.0.0
+updated: 2026-07-03
 description: Tải các file liên quan và vẽ biểu đồ phụ thuộc (dependency graph) cho một tác vụ.
 category: utilities
-behavior: static-analysis
-intent: review-code
-priority: medium
 tags: [context, filesystem, dependencies, graph]
 platforms: [claude-code, cursor, windsurf, gemini-cli]
-trigger: AI needs to understand the surrounding files before modifying a core component.
-inputs: [Target File]
-outputs: [Loaded Context, Dependency Graph]
-allowed_tools: [grep_search, read_file, list_dir]
-pipeline: [analyze, implement, complete]
 ---
 
-# 🛠️ qk-context-loader - Standard Operating Procedure
+# 🛠️ qk-context-loader
 
-> **Goal:** "Người thu thập tình báo". Đảm bảo AI hiểu rõ môi trường xung quanh một file code trước khi tiến hành sửa nó, tránh lỗi thiếu context.
+> **Inheritance:** Kỹ năng này tuân thủ Kiến trúc v5.0 của `framework/KERNEL.md`.
+> Output bắt buộc là Decision Summary.
 
-## 🔄 1. Chain of Thought (SOP)
+---
 
-1. **Analyze (Target Identification):**
-   - Identify the main file that needs modification.
-2. **Implement (Graph Traversal):**
-   - **Load Parents:** Search for files that import the target file.
-   - **Load Children:** Read the target file to see what it imports.
-   - Gather the essential chunks of these related files.
-3. **Complete (Context Mapping):**
-   - Build a mental map of how the data flows.
+## 🎯 Mission (Scope)
+- ✅ Tìm và nạp các file phụ thuộc (Dependencies, Imports).
+- ❌ Do NOT nạp toàn bộ repo hoặc thư mục `node_modules` gây phình Context Window.
 
-## 🛡️ 2. Constraints & Rules
+---
 
-- **Token Optimization:** Do not load `node_modules` or massive minified build files. Extract only the exact logic needed.
+## ⚙️ Capabilities (Cognitive Pipeline)
+```yaml
+Pipeline:
+  - inference
+  - planning
+  - execution
+  - bias-review
+  - ship-check
+```
 
-## 🌳 3. Decision Tree
+---
 
-```text
-Is the target file a UI Component?
-  ├── YES → Load its CSS/Theme tokens and any nested Child components.
-  └── NO (API Service) → Load the corresponding Controller and Database Model.
+## 🎛️ Dials (Hành vi)
+```yaml
+Dials:
+  - id: complexity-budget
 ```
 
-## 🤝 4. Handoff Pipeline
-
-1. `complete`: Provide the context map to the Orchestrator or Target Skill.
+---
 
-## 📝 5. Output Format
+## 🛡️ Biases (Sửa lỗi mặc định)
+```yaml
+Biases:
+  - id: hallucinated-tools # (Đoán mò tên file thay vì tìm kiếm thật)
+```
 
-Vui lòng trả kết quả bằng Tiếng Việt.
+---
 
-- **Tóm tắt (Summary):** Đã nạp thành công ngữ cảnh.
-- **Chi tiết (Changes):** Các file có liên quan trực tiếp.
+## 🛫 Ship Criteria
+```yaml
+Rules:
+  - id: delegation-only # (Chỉ thu thập thông tin, cấm chỉnh sửa code)
+```

package/skills/qk-data-lifecycle/SKILL.md

@@ -1,68 +1,56 @@
 ---
 name: qk-data-lifecycle
-version: 3.1.5
-updated: 2026-07-02
+version: 5.0.0
+updated: 2026-07-03
 description: Quản lý Database Schema, Migrations, và Repositories.
 category: database
-behavior: development
-intent: implement-feature
-priority: medium
-tags: [database, sql, migration, schema, orm]
+tags: [database, schema, migrations, orm]
 platforms: [claude-code, cursor, windsurf, gemini-cli]
-trigger: User asks to create a new database table, modify schema, or write complex queries.
-inputs: [Data model requirements]
-outputs: [Migration scripts, Models, Repositories]
-allowed_tools: [run_command, read_file, write_to_file]
-pipeline: [analyze, design, implement, engineering-standard, validate, complete]
 ---
 
-# 🛠️ qk-data-lifecycle - Standard Operating Procedure
+# 🛠️ qk-data-lifecycle
 
-> **Goal:** Quản lý an toàn vòng đời dữ liệu, đảm bảo Schema chuẩn xác, Migration có thể Rollback và Query tối ưu.
+> **Inheritance:** Kỹ năng này tuân thủ Kiến trúc v5.0 của `framework/KERNEL.md`.
+> Output bắt buộc là Decision Summary.
 
-## 🔄 1. Chain of Thought (SOP)
+---
 
-1. **Analyze (Requirements):**
-   - Understand the entities and their relationships (1:1, 1:N, N:N).
-   - Identify the ORM/Query Builder used in the project (e.g., Prisma, TypeORM, Drizzle).
-2. **Design (Schema & Normalization):**
-   - Design the schema. Ensure foreign keys and indices are applied to frequently queried columns.
-   - Always include audit columns (`created_at`, `updated_at`).
-3. **Implement (Migration & Models):**
-   - Write the Migration script. MUST include both `up` (apply) and `down` (rollback) logic.
-   - Write/Update the Model/Entity files.
-   - Create Repository layer functions if necessary.
-4. **Verify (Safety Check):**
-   - Review the migration script for destructive operations (e.g., `DROP TABLE`, `DROP COLUMN`).
+## 🎯 Mission (Scope)
+- ✅ Thiết kế cấu trúc CSDL và viết file Migration an toàn.
+- ❌ Do NOT xóa cột hoặc đổi tên cột đang dùng (DROP/RENAME) mà không dùng cơ chế "Expand & Contract".
 
-## 🛡️ 2. Constraints & Rules
+---
 
-- **No Data Loss:** NEVER execute a migration that drops data on Production without explicit User approval and a backup plan.
-- **Decoupling:** Do not write SQL queries directly in Controllers. Always use the Repository or DAO pattern.
-- **Index Rule:** Any column used in a `WHERE`, `JOIN`, or `ORDER BY` clause should be evaluated for an index.
+## ⚙️ Capabilities (Cognitive Pipeline)
+```yaml
+Pipeline:
+  - inference
+  - planning
+  - execution
+  - bias-review
+  - ship-check
+```
 
-## 🌳 3. Decision Tree
+---
 
-```text
-Is this modifying an existing table in Production?
-  ├── YES → Does it drop a column or change a data type destructively?
-  │       ├── YES → STOP. Create a new column, migrate data, then drop the old one (Expand & Contract Pattern).
-  │       └── NO → Proceed with safe migration (e.g., ADD COLUMN).
-  └── NO (New Table) → Ensure Primary Key, Timestamps, and Foreign Keys are defined.
+## 🎛️ Dials (Hành vi)
+```yaml
+Dials:
+  - id: strictness
 ```
 
-## 🤝 4. Handoff Pipeline
-
-1. `engineering-standard`: Verify naming conventions (e.g., snake_case for DB columns if required).
-2. `validate`: Run dry-run migrations or unit tests.
-3. `complete`: Generate the Database report.
+---
 
-## 📝 5. Output Format
+## 🛡️ Biases (Sửa lỗi mặc định)
+```yaml
+Biases:
+  - id: over-indexing # (Chống thói quen đánh index bừa bãi)
+```
 
-Vui lòng trả kết quả bằng Tiếng Việt.
+---
 
-- **Tóm tắt (Summary):** Bảng nào được tạo/sửa.
-- **Chi tiết (Changes):** File migration và Model đã sinh ra.
-- **Kiến trúc (Reasoning):** Lý do thiết kế quan hệ bảng và các Index.
-- **Rủi ro (Risks):** Đánh giá mức độ an toàn của Migration (Có làm mất data không?).
-- **Hành động tiếp (Next Action):** Hướng dẫn user chạy lệnh migrate.
+## 🛫 Ship Criteria
+```yaml
+Rules:
+  - id: minimal-diff
+```

package/skills/qk-db-optimizer/SKILL.md

@@ -1,63 +1,56 @@
 ---
 name: qk-db-optimizer
-version: 1.0.0
-updated: 2026-07-02
+version: 5.0.0
+updated: 2026-07-03
 description: Tối ưu hóa Database (đánh Index, xử lý N+1 Queries, Slow Queries).
 category: database
-behavior: development
-intent: implement-feature
-priority: high
-tags: [database, optimize, sql, index, performance]
+tags: [database, performance, optimization, queries]
 platforms: [claude-code, cursor, windsurf, gemini-cli]
-trigger: User asks to optimize database, fix slow queries, add index, or resolve N+1 issues.
-inputs: [Database Schema, Slow Query Log or ORM code]
-outputs: [Optimized Query/ORM code, Migration script for Indexes]
-allowed_tools: [run_command, read_file, write_to_file, grep_search]
-pipeline: [analyze, plan, implement, validate, complete]
 ---
 
-# 🛠️ qk-db-optimizer - Standard Operating Procedure
+# 🛠️ qk-db-optimizer
 
-> **Goal:** Tối ưu hóa hiệu suất cơ sở dữ liệu. Đóng vai trò là một Database Administrator (DBA) / Performance Engineer. Giải quyết các bài toán cổ điển như N+1 queries, thiếu Index, truy vấn chậm.
+> **Inheritance:** Kỹ năng này tuân thủ Kiến trúc v5.0 của `framework/KERNEL.md`.
+> Output bắt buộc là Decision Summary.
 
-## 🔄 1. Chain of Thought (SOP)
+---
 
-1. **Analyze (Bottleneck Identification):**
-   - Read the relevant ORM code (Prisma, TypeORM, Drizzle) or raw SQL.
-   - Identify the source of the slowdown (e.g., fetching in a loop causing N+1).
-2. **Plan (Optimization Strategy):**
-   - Decide whether to optimize the query structure (e.g., using `JOIN` / `include`) or add an Index to the database schema.
-3. **Implement (Code & Schema changes):**
-   - Refactor the code to eliminate N+1 queries.
-   - Write migration scripts or update the ORM schema to add necessary Indexes.
-4. **Verify (Explain & Test):**
-   - Ensure the new query yields the same correct result.
+## 🎯 Mission (Scope)
+- ✅ Tìm nguyên nhân gây nghẽn cổ chai DB và đề xuất Query Optimization.
+- ❌ Do NOT vội vàng đề xuất Redis/Cache trước khi fix tận gốc câu SQL lỗi.
 
-## 🛡️ 2. Constraints & Rules
+---
 
-- **No Destructive Operations:** Never drop tables or columns unless explicitly asked by the user to do so.
-- **Explain First:** Always explain why a query is slow and how your index/refactor fixes it.
+## ⚙️ Capabilities (Cognitive Pipeline)
+```yaml
+Pipeline:
+  - inference
+  - planning
+  - execution
+  - bias-review
+  - ship-check
+```
 
-## 🌳 3. Decision Tree
+---
 
-```text
-Is it an N+1 Query problem?
-  ├── YES → Use Eager Loading (e.g., `.include()` in Prisma) or a DataLoader pattern.
-  └── NO → Is it a slow filtering/sorting issue?
-             ├── YES → Add a database Index (B-Tree, Hash, etc.) to the relevant columns.
-             └── NO → Refactor the raw SQL/ORM logic to be more efficient.
+## 🎛️ Dials (Hành vi)
+```yaml
+Dials:
+  - id: optimization-strategy
 ```
 
-## 🤝 4. Handoff Pipeline
-
-1. `validate`: Run tests using `qk-validation-gate` to ensure no data regressions.
-2. `complete`: Generate the Database Optimization report.
+---
 
-## 📝 5. Output Format
+## 🛡️ Biases (Sửa lỗi mặc định)
+```yaml
+Biases:
+  - id: over-indexing
+```
 
-Vui lòng trả kết quả bằng Tiếng Việt.
+---
 
-- **Tóm tắt (Summary):** Vấn đề gì đã được giải quyết.
-- **Chi tiết (Changes):** Cách khắc phục (đánh index hay refactor code).
-- **Hiệu năng (Performance):** Giải thích tại sao cách mới lại nhanh hơn.
-- **Xác thực (Verification):** ...
+## 🛫 Ship Criteria
+```yaml
+Rules:
+  - id: minimal-diff
+```