validate-album
정보
이 스킬은 Bash, Grep, 음악 전용 MCP와 같은 도구를 사용하여 파일 위치와 콘텐츠 무결성을 확인함으로써 음악 앨범 디렉토리 구조를 검증합니다. 출시 전 품질 보증 또는 개발자가 앨범 구조 상태를 확인해야 할 때 사용하도록 설계되었습니다. 이 스킬은 자동으로 구성을 로드하고 퍼지 매칭을 통해 앨범을 찾은 후, 누락된 파일이나 경로 문제를 보고합니다.
빠른 설치
Claude Code
추천npx skills add bitwize-music-studio/claude-ai-music-skills -a claude-code/plugin add https://github.com/bitwize-music-studio/claude-ai-music-skillsgit clone https://github.com/bitwize-music-studio/claude-ai-music-skills.git ~/.claude/skills/validate-albumClaude Code에서 이 명령을 복사하여 붙여넣어 스킬을 설치하세요
문서
Album Validator Agent
Your Task
Input: $ARGUMENTS (album name, e.g., sample-album)
Validate that an album has all required files in the correct locations, catching path issues and missing content before they become problems.
Step 1: Load Config & Find Album
-
Call
get_config()— returns paths (content_root,audio_root,documents_root) andartist.name- If config missing, STOP and report:
[FAIL] Config file missing: ~/.bitwize-music/config.yaml Run /configure to set up the plugin.
- If config missing, STOP and report:
-
Call
find_album(album_name)— fuzzy match by name, slug, or partial- If not found, STOP and report (MCP returns available albums):
[FAIL] Album not found: {album-name}
- If not found, STOP and report (MCP returns available albums):
-
Optionally call
validate_album_structure(album_slug)— runs structural validation checks and returns{passed, failed, warnings, skipped, issues[], checks[]}. This MCP tool handles directory structure, required files, audio placement, and track content checks in one call.
Note: The MCP validate_album_structure tool performs many of the checks below automatically. You can use its results directly or run the manual checks for more detailed reporting.
Step 3: Run Validations
Initialize Counters
passed = 0failed = 0warnings = 0skipped = 0issues = [](list of fix commands)
Output Header
═══════════════════════════════════════════════════════════
ALBUM VALIDATION: {album-name}
═══════════════════════════════════════════════════════════
Validation Categories
CONFIG
CONFIG
──────
| Check | Pass | Fail |
|---|---|---|
| Config file exists | [PASS] Config file exists | [FAIL] Config file missing |
| content_root defined | [PASS] content_root: {value} | [FAIL] content_root not defined |
| audio_root defined | [PASS] audio_root: {value} | [FAIL] audio_root not defined |
| artist defined | [PASS] artist: {value} | [FAIL] artist.name not defined |
ALBUM STRUCTURE
ALBUM STRUCTURE
───────────────
| Check | How | Pass | Fail |
|---|---|---|---|
| Album dir exists | test -d {album_path} | [PASS] Album directory: {path} | [FAIL] Album directory missing |
| README.md exists | test -f {album_path}/README.md | [PASS] README.md exists | [FAIL] README.md missing |
| tracks/ dir exists | test -d {album_path}/tracks | [PASS] tracks/ directory exists | [FAIL] tracks/ directory missing |
| Track files exist | ls {album_path}/tracks/*.md | [PASS] {N} track files found | [WARN] No track files found |
For documentary albums (check README.md for type):
| Check | How | Pass | Fail |
|---|---|---|---|
| RESEARCH.md exists | test -f {album_path}/RESEARCH.md | [PASS] RESEARCH.md exists | [WARN] RESEARCH.md missing (documentary album) |
| SOURCES.md exists | test -f {album_path}/SOURCES.md | [PASS] SOURCES.md exists | [WARN] SOURCES.md missing (documentary album) |
AUDIO FILES
AUDIO FILES
───────────
Expected path: {audio_root}/artists/{artist}/albums/{genre}/{album}/
| Check | How | Pass | Fail |
|---|---|---|---|
| Audio dir exists (correct path) | test -d {audio_root}/artists/{artist}/albums/{genre}/{album} | [PASS] Audio directory: {path} | See below |
| Audio dir in wrong location | test -d {audio_root}/{album} | N/A | [FAIL] Audio in wrong location (missing artist folder) |
If audio in wrong location, add to issues:
→ Expected: {audio_root}/artists/{artist}/albums/{genre}/{album}/
→ Found at: {audio_root}/{album}/ (WRONG - missing artist folder)
→ Fix: mv {audio_root}/{album}/ {audio_root}/artists/{artist}/albums/{genre}/{album}/
| Check | How | Pass | Skip |
|---|---|---|---|
| WAV files present | ls {audio_path}/*.wav | [PASS] {N} WAV files found | [SKIP] No audio files yet |
| mastered/ exists | test -d {audio_path}/mastered | [PASS] mastered/ directory exists | [SKIP] Not mastered yet |
ALBUM ART
ALBUM ART
─────────
| Check | How | Pass | Skip |
|---|---|---|---|
| Art in audio folder | test -f {audio_path}/album.png | [PASS] album.png in audio folder | [SKIP] No album art yet |
| Art in content folder | test -f {album_path}/album-art.* | [PASS] album-art in content folder | [SKIP] No album art yet |
TRACKS
TRACKS
──────
For each track file in {album_path}/tracks/*.md:
- Read the file
- Check for required fields:
- Status field exists
- Suno Style Box exists (has
## Suno Inputssection) - Suno Lyrics Box exists
- If Status is
GeneratedorFinal: Suno Link present - If documentary: Sources Verified status
- Check instrumental field sync:
- Read frontmatter
instrumentalfield (true/false/missing) - Read Track Details table
**Instrumental**row (Yes/No/missing) - If both present and they disagree →
[WARN] {filename} - Instrumental field mismatch: frontmatter={value}, table={value} - If only one is set →
[WARN] {filename} - Instrumental field missing from {frontmatter|table} (set in {other})
- Read frontmatter
Output per track:
[PASS] {filename} - Status: {status}, Suno Link: {present/missing}[WARN] {filename} - Status: {status}, missing {what}[FAIL] {filename} - No Status field
Step 4: Summary
═══════════════════════════════════════════════════════════
SUMMARY: {passed} passed, {failed} failed, {warnings} warning(s), {skipped} skipped
═══════════════════════════════════════════════════════════
If any issues:
ISSUES TO FIX:
1. {issue description}
{fix command}
2. ...
Example Output
═══════════════════════════════════════════════════════════
ALBUM VALIDATION: sample-album
═══════════════════════════════════════════════════════════
CONFIG
──────
[PASS] Config file exists
[PASS] content_root: ~/bitwize-music
[PASS] audio_root: ~/bitwize-music/audio
[PASS] artist: bitwize
ALBUM STRUCTURE
───────────────
[PASS] Album directory: ~/bitwize-music/artists/bitwize/albums/electronic/sample-album/
[PASS] README.md exists
[PASS] tracks/ directory exists
[PASS] 5 track files found
AUDIO FILES
───────────
[FAIL] Audio directory in wrong location
→ Expected: ~/bitwize-music/audio/artists/bitwize/albums/electronic/sample-album/
→ Found at: ~/bitwize-music/audio/sample-album/
→ Fix: mv ~/bitwize-music/audio/sample-album/ ~/bitwize-music/audio/artists/bitwize/albums/electronic/sample-album/
ALBUM ART
─────────
[SKIP] No album art yet
TRACKS
──────
[PASS] 01-intro.md - Status: Final, Suno Link: present
[PASS] 02-track.md - Status: Final, Suno Link: present
[WARN] 03-t-day-beach.md - Status: Generated, Suno Link: missing
═══════════════════════════════════════════════════════════
SUMMARY: 8 passed, 1 failed, 1 warning, 1 skipped
═══════════════════════════════════════════════════════════
ISSUES TO FIX:
1. Move audio folder to include artist:
mv ~/bitwize-music/audio/sample-album/ ~/bitwize-music/audio/artists/bitwize/albums/electronic/sample-album/
Important Notes
- Use MCP tools first -
get_config(),find_album(),validate_album_structure()before manual checks - Check both correct AND wrong locations - Catch misplaced files
- Provide actionable fixes - Include exact commands to fix issues
- Use appropriate status - PASS/FAIL/WARN/SKIP based on severity
- Count everything - Report totals in summary
GitHub 저장소
연관 스킬
llamaguard
기타LlamaGuard는 폭력 및 혐오 발언 등 6가지 안전 범주에서 LLM 입력과 출력을 조정하기 위한 Meta의 70-80억 파라미터 모델입니다. 94-95% 정확도를 제공하며 vLLM, Hugging Face 또는 Amazon SageMaker를 사용해 배포할 수 있습니다. 이 기술을 사용하여 AI 애플리케이션에 콘텐츠 필터링 및 안전 가드레일을 손쉽게 통합하세요.
cost-optimization
기타이 Claude Skill은 리소스 적정화, 태깅 전략, 지출 분석을 통해 개발자들이 클라우드 비용을 최적화할 수 있도록 지원합니다. AWS, Azure, GCP에서 클라우드 비용을 절감하고 비용 거버넌스를 구현하기 위한 프레임워크를 제공합니다. 인프라 비용을 분석하거나, 리소스를 적정화하거나, 예산 제약을 충족해야 할 때 사용하세요.
quantizing-models-bitsandbytes
기타이 스킬은 bitsandbytes를 사용하여 LLM을 8비트 또는 4비트 정밀도로 양자화하며, 최소한의 정확도 손실로 50-75%의 메모리 감소를 달성합니다. 제한된 GPU 메모리에서 더 큰 모델을 실행하거나 추론을 가속화하는 데 이상적이며, INT8, NF4, FP4와 같은 형식을 지원합니다. 이 스킬은 HuggingFace Transformers와 통합되어 QLoRA 학습 및 8비트 옵티마이저를 가능하게 합니다.
dispatching-parallel-agents
기타이 Claude Skill은 3개 이상의 독립적인 문제를 동시에 조사하고 해결하기 위해 다중 에이전트를 배치합니다. 공유 상태나 의존성 없이 해결 가능한 무관련 장애 시나리오에 맞게 설계되었습니다. 핵심 기능은 병렬 문제 해결로, 각 독립 문제 영역마다 하나의 에이전트를 할당하여 효율성을 극대화합니다.
