# Send Quality-Driven Development (QDD) to your agent
Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.
## Fast path
- Download the package from Yavira.
- Extract it into a folder your agent can access.
- Paste one of the prompts below and point your agent at the extracted folder.
## Suggested prompts
### New install

```text
I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete.
```
### Upgrade existing

```text
I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run.
```
## Machine-readable fields
```json
{
  "schemaVersion": "1.0",
  "item": {
    "slug": "quality-driven-dev",
    "name": "Quality-Driven Development (QDD)",
    "source": "tencent",
    "type": "skill",
    "category": "效率提升",
    "sourceUrl": "https://clawhub.ai/kimky1122/quality-driven-dev",
    "canonicalUrl": "https://clawhub.ai/kimky1122/quality-driven-dev",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadUrl": "/downloads/quality-driven-dev",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=quality-driven-dev",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "packageFormat": "ZIP package",
    "primaryDoc": "SKILL.md",
    "includedAssets": [
      "SKILL.md",
      "references/ddd-patterns.md",
      "references/tdd-patterns.md",
      "references/trust5-checklist.md"
    ],
    "downloadMode": "redirect",
    "sourceHealth": {
      "source": "tencent",
      "slug": "quality-driven-dev",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-05-01T01:08:14.193Z",
      "expiresAt": "2026-05-08T01:08:14.193Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=quality-driven-dev",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=quality-driven-dev",
        "contentDisposition": "attachment; filename=\"quality-driven-dev-1.1.0.zip\"",
        "redirectLocation": null,
        "bodySnippet": null,
        "slug": "quality-driven-dev"
      },
      "scope": "item",
      "summary": "Item download looks usable.",
      "detail": "Yavira can redirect you to the upstream package for this item.",
      "primaryActionLabel": "Download for OpenClaw",
      "primaryActionHref": "/downloads/quality-driven-dev"
    },
    "validation": {
      "installChecklist": [
        "Use the Yavira download entry.",
        "Review SKILL.md after the package is downloaded.",
        "Confirm the extracted package contains the expected setup assets."
      ],
      "postInstallChecks": [
        "Confirm the extracted package includes the expected docs or setup files.",
        "Validate the skill or prompts are available in your target agent workspace.",
        "Capture any manual follow-up steps the agent could not complete."
      ]
    }
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/quality-driven-dev",
    "downloadUrl": "https://openagent3.xyz/downloads/quality-driven-dev",
    "agentUrl": "https://openagent3.xyz/skills/quality-driven-dev/agent",
    "manifestUrl": "https://openagent3.xyz/skills/quality-driven-dev/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/quality-driven-dev/agent.md"
  }
}
```
## Documentation

### Quality-Driven Development

Structured development methodology inspired by MoAI-ADK's TRUST 5 framework. Automatically selects TDD or DDD based on project state, enforces quality gates, and produces tested, documented code.

### Core Philosophy

"바이브 코딩의 목적은 빠른 생산성이 아니라 코드 품질이다."

### Logging Strategy

All code must include meaningful logs. Logs are the first line of defense for debugging production issues.

### Log Levels

LevelPurposeExamples운영(PRD)개발(DEV)ERROR예외, 실패, 복구 불가 상황catch 블록, DB 연결 실패, 필수값 누락✅✅WARN예상 밖 상황, 복구 가능fallback 사용, 재시도, deprecated 호출✅✅INFO핵심 흐름만 간결하게API 호출/응답, 상태 변경, 트랜잭션 시작/완료✅✅DEBUG상세 디버깅, 자유롭게함수 진입/종료, 변수값, 조건 분기, 쿼리 파라미터❌✅

### Log Placement Rules

반드시 로그를 넣어야 하는 곳:

API 엔드포인트 진입 (INFO: 요청 파라미터 요약)
외부 서비스 호출 전/후 (INFO: 호출 대상, 응답 상태)
에러/예외 catch 블록 (ERROR: 에러 메시지 + 컨텍스트)
비즈니스 로직 분기점 (DEBUG: 어떤 분기로 갔는지)
상태 변경 (INFO: before → after)
배치/스케줄러 시작/완료 (INFO: 처리 건수, 소요 시간)

로그 작성 원칙:

운영에서 INFO만으로 흐름 추적이 가능해야 한다
DEBUG는 부담 없이 자유롭게 — 운영에선 출력 안 됨
민감 정보(비밀번호, 토큰, 개인정보) 절대 로그에 포함 금지
로그 메시지에 컨텍스트 포함 (ID, 파라미터 등) — "처리 실패" ❌ → "주문 처리 실패 [orderId=123, reason=재고부족]" ✅

### Phase 0: Project Analysis

Before any coding, analyze the project:

Check if test framework exists (jest, vitest, pytest, go test, etc.)
Measure current test coverage (run coverage command if available)
Detect language, framework, and project structure
Identify logging framework (slf4j, winston, pino, logback, print/console.log etc.) — if none exists, recommend and set up one
Select methodology automatically:

Coverage >= 10% OR new project → TDD (default)
Coverage < 10% AND existing project → DDD

Report the analysis result and selected methodology to the user before proceeding.

### Phase 1: SPEC Document

Create a SPEC document before implementation:

# SPEC-{ID}: {Title}

## Goal
One sentence describing what this change achieves.

## Acceptance Criteria
- [ ] Criterion 1 (testable)
- [ ] Criterion 2 (testable)
- [ ] Criterion 3 (testable)

## Scope
- **In scope:** What will be changed
- **Out of scope:** What will NOT be changed

## Technical Approach
Brief description of implementation strategy.

## Log Points
Key locations where logs will be added (level + message summary).

## TRUST 5 Checklist
- [ ] **Tested:** All acceptance criteria have corresponding tests
- [ ] **Readable:** Code is self-documenting with clear naming
- [ ] **Unified:** Follows existing project conventions
- [ ] **Secured:** No new vulnerabilities introduced
- [ ] **Trackable:** Changes are documented and linked to this SPEC

### Phase 2A: TDD Execution (New Projects / Coverage >= 10%)

Follow RED → GREEN → REFACTOR strictly:

RED — Write failing tests first

Write test for first acceptance criterion
Run test — confirm it FAILS
Report: "🔴 RED: Test written and failing as expected"

GREEN — Minimal implementation

Write minimum code to pass the test
Add appropriate logs at key points (API calls, error handling, state changes)
Run test — confirm it PASSES
Report: "🟢 GREEN: Test passing"

REFACTOR — Clean up

Improve code quality while keeping tests green
Review log quality — ensure levels are correct, messages are clear with context
Run all tests — confirm everything still passes
Report: "♻️ REFACTOR: Code cleaned, all tests green"

Repeat for each acceptance criterion.

### Phase 2B: DDD Execution (Existing Projects / Coverage < 10%)

Follow ANALYZE → PRESERVE → IMPROVE:

ANALYZE — Understand existing code

Read existing code and identify dependencies
Map domain boundaries and side effects
Check existing logging — identify gaps where logs are missing
Report: "🔍 ANALYZE: Current behavior documented"

PRESERVE — Capture current behavior

Write characterization tests for existing behavior
Run tests — confirm they pass against current code
Report: "🛡️ PRESERVE: Characterization tests in place"

IMPROVE — Change under test protection

Make changes incrementally
Add/improve logs at changed code paths
Run tests after each change
Report: "📈 IMPROVE: Changes verified by tests"

### Phase 3: TRUST 5 Quality Gate

Before declaring work complete, verify all 5 principles:

PrincipleCheckActionTestedRun full test suiteAll tests pass, coverage maintained or improvedReadableReview naming, comments, log messagesFix unclear names, ensure log messages have contextUnifiedCheck style consistency, log format consistencyMatch existing patterns (indent, naming, log format)SecuredSecurity review, log content reviewNo hardcoded secrets, no sensitive data in logsTrackableDocumentation, log coverageChanges described, key paths have appropriate logs

Only proceed to completion when ALL 5 checks pass.

### Phase 4: Completion Report

## ✅ SPEC-{ID} Complete

### Methodology: {TDD|DDD}
### Changes:
- {file1}: {what changed}
- {file2}: {what changed}

### Log Points Added:
- {file1:line}: {level} - {description}
- {file2:line}: {level} - {description}

### Test Results:
- Tests: {passed}/{total}
- Coverage: {before}% → {after}%

### TRUST 5:
- ✅ Tested | ✅ Readable | ✅ Unified | ✅ Secured | ✅ Trackable

### Agent Roles

When working on complex tasks, delegate to specialized perspectives:

RoleFocusWhen to ActivateArchitectSystem design, API contractsNew feature, structural changeBackendAPI, DB, business logicServer-side workFrontendUI, UX, componentsClient-side workSecurityVulnerabilities, auth, input validationAuth features, data handlingTesterTest strategy, edge cases, coverageAlways (TRUST 5 - Tested)PerformanceOptimization, profilingLoad-sensitive features

For each task, identify which roles are relevant and apply their perspective during review.

### Reference Guides

TopicReferenceLoad WhenTDD Patternsreferences/tdd-patterns.mdTDD methodology selectedDDD Patternsreferences/ddd-patterns.mdDDD methodology selectedTRUST 5 Detailreferences/trust5-checklist.mdQuality gate phaseLanguage-specificreferences/lang-{language}.mdLanguage-specific patterns needed

### Constraints

MUST DO:

Always analyze project before choosing methodology
Always create SPEC before coding
Always write tests (TDD: before code, DDD: before changes)
Always run TRUST 5 gate before completion
Report progress at each phase transition
Always add meaningful logs with appropriate levels at key code points
Always ensure tests are actually executed (not just written) — run the test suite and confirm results before proceeding

MUST NOT:

Skip test writing for any reason
Write implementation before tests (TDD mode)
Modify untested code without characterization tests first (DDD mode)
Declare complete without all 5 TRUST checks passing
Change code outside the SPEC scope
Log sensitive data (passwords, tokens, personal info)
Skip logging at error/catch blocks
## Trust
- Source: tencent
- Verification: Indexed source record
- Publisher: kimky1122
- Version: 1.1.0
## Source health
- Status: healthy
- Item download looks usable.
- Yavira can redirect you to the upstream package for this item.
- Health scope: item
- Reason: direct_download_ok
- Checked at: 2026-05-01T01:08:14.193Z
- Expires at: 2026-05-08T01:08:14.193Z
- Recommended action: Download for OpenClaw
## Links
- [Detail page](https://openagent3.xyz/skills/quality-driven-dev)
- [Send to Agent page](https://openagent3.xyz/skills/quality-driven-dev/agent)
- [JSON manifest](https://openagent3.xyz/skills/quality-driven-dev/agent.json)
- [Markdown brief](https://openagent3.xyz/skills/quality-driven-dev/agent.md)
- [Download page](https://openagent3.xyz/downloads/quality-driven-dev)