gsd-skill-creator MCP — 사용 사례, 설치 & 라이브 데모

왜 쓰나요

핵심 기능

Six-step adaptive cycle: observe, detect, suggest, apply, learn, compose
3-repetition threshold before suggesting a skill — avoids noise
Learning guardrails: 3-correction minimum, 7-day cooldown
GSD-OS desktop app (Tauri, retro UI) for managing the framework
Chipset architecture for resource coordination (Amiga-inspired)

라이브 데모

실제 사용 모습

gsd-skill-creator-skill.replay ▶ 준비됨

0/0

설치

클라이언트 선택

~/Library/Application Support/Claude/claude_desktop_config.json · Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "gsd-skill-creator-skill": {
      "command": "git",
      "args": [
        "clone",
        "https://github.com/Tibsfox/gsd-skill-creator",
        "~/.claude/skills/gsd-skill-creator"
      ],
      "_inferred": true
    }
  }
}

Claude Desktop → Settings → Developer → Edit Config 열기. 저장 후 앱 재시작.

~/.cursor/mcp.json · .cursor/mcp.json

{
  "mcpServers": {
    "gsd-skill-creator-skill": {
      "command": "git",
      "args": [
        "clone",
        "https://github.com/Tibsfox/gsd-skill-creator",
        "~/.claude/skills/gsd-skill-creator"
      ],
      "_inferred": true
    }
  }
}

Cursor는 Claude Desktop과 동일한 mcpServers 스키마 사용. 프로젝트 설정이 전역보다 우선.

VS Code → Cline → MCP Servers → Edit

{
  "mcpServers": {
    "gsd-skill-creator-skill": {
      "command": "git",
      "args": [
        "clone",
        "https://github.com/Tibsfox/gsd-skill-creator",
        "~/.claude/skills/gsd-skill-creator"
      ],
      "_inferred": true
    }
  }
}

Cline 사이드바의 MCP Servers 아이콘 클릭 후 "Edit Configuration" 선택.

~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "gsd-skill-creator-skill": {
      "command": "git",
      "args": [
        "clone",
        "https://github.com/Tibsfox/gsd-skill-creator",
        "~/.claude/skills/gsd-skill-creator"
      ],
      "_inferred": true
    }
  }
}

Claude Desktop과 같은 형식. Windsurf 재시작 후 적용.

~/.continue/config.json

{
  "mcpServers": [
    {
      "name": "gsd-skill-creator-skill",
      "command": "git",
      "args": [
        "clone",
        "https://github.com/Tibsfox/gsd-skill-creator",
        "~/.claude/skills/gsd-skill-creator"
      ]
    }
  ]
}

Continue는 맵이 아닌 서버 오브젝트 배열 사용.

~/.config/zed/settings.json

{
  "context_servers": {
    "gsd-skill-creator-skill": {
      "command": {
        "path": "git",
        "args": [
          "clone",
          "https://github.com/Tibsfox/gsd-skill-creator",
          "~/.claude/skills/gsd-skill-creator"
        ]
      }
    }
  }
}

context_servers에 추가. 저장 시 Zed가 핫 리로드.

claude mcp add gsd-skill-creator-skill -- git clone https://github.com/Tibsfox/gsd-skill-creator ~/.claude/skills/gsd-skill-creator

한 줄 명령. claude mcp list로 확인, claude mcp remove로 제거.

사용 사례

실전 활용법: gsd-skill-creator

Let GSD detect your repeating workflow and codify it as a skill

👤 Heavy Claude Code users who keep redoing the same steps ⏱ ~30 min intermediate

언제 쓸까: You've done 'clone repo -> run tests -> fix lint -> open PR' enough times that muscle memory should be code.

사전 조건

Skill cloned — git clone https://github.com/Tibsfox/gsd-skill-creator ~/.claude/skills/gsd-skill-creator

흐름

Run with observation mode on

Enable gsd-skill-creator in observe mode. Run my normal PR flow a few times over the next week.✓ 복사됨

→ Silent logging of tool sequences
Review suggestions

Show me detected patterns with 3+ occurrences, ranked by confidence.✓ 복사됨

→ Ranked list with proposed skill names
Confirm and let it build

Promote pattern #1 into a skill — approve generation.✓ 복사됨

→ A new skill appears in ~/.claude/skills/

결과: Your actual workflow becomes a reusable skill, grounded in data.

함정

Approving every suggestion — Promote only the top 1–2 — skill bloat hurts the picker

함께 쓰기: skill-forge-skill

Compose co-activating skills into an agent team

👤 Engineers whose skills consistently fire together ⏱ ~30 min advanced

언제 쓸까: After a month of use, the skill picker keeps selecting the same 3 skills in sequence.

흐름

Review co-activation report

Show skill co-activation report — which skills fire within the same task 80%+ of the time?✓ 복사됨

→ Cluster diagram
Compose team

Compose the top cluster into a named agent team.✓ 복사됨

→ Team definition that orchestrates the cluster

결과: A higher-level orchestrator that triggers the full cluster in one shot.

Resume a long project without repeating old mistakes

👤 Teams doing multi-week Claude-driven work ⏱ ~15 min beginner

언제 쓸까: You're picking up a project Claude has lost context on.

흐름

Resume with skill history

Resume project X using gsd-skill-creator memory — what patterns and corrections have I applied so far?✓ 복사됨

→ Summary of prior skills and corrections

결과: Claude starts where it left off, not from zero.

함께 쓰기: planning-with-files-skill

조합

다른 MCP와 조합해 10배 효율

gsd-skill-creator-skill + skill-forge-skill

GSD detects the pattern; skill-forge scaffolds the formal skill

Hand the top detected pattern to skill-forge build for proper structure.✓ 복사됨

gsd-skill-creator-skill + planning-with-files-skill

GSD observes sessions; planning-with-files persists state across them

Use both — GSD for pattern learning, planning-with-files for session state.✓ 복사됨

도구

이 MCP가 노출하는 것

도구	입력	언제 호출
observe_start	session id or none	Start of the adoption period
detect_patterns	min_occurrences?	Weekly review
promote_to_skill	pattern id, name	After user approval
compose_team	cluster id	When clusters are stable
learn_from_correction	correction context	After 3+ corrections in cooldown window

비용 및 제한

운영 비용

API 쿼터: None
호출당 토큰: Observation is cheap; generation phases are moderate
금액: Free, BSL 1.1 (converts to GPL 3.0 in 2030)
팁: Only run detection weekly, not per-session

보안

권한, 시크릿, 파급범위

자격 증명 저장: Observation data is local

데이터 외부 송신: None by default

GSD writes generated skills into ~/.claude/skills/ — review them before trusting.

문제 해결

자주 발생하는 오류와 해결

No patterns detected after weeks of use

Your sessions may vary too much; loosen the pattern-matching threshold

GSD-OS desktop app won't launch

Rebuild Tauri bundle for your platform; check Rust toolchain

확인: cargo --version

대안

gsd-skill-creator 다른 것과 비교

대안	언제 쓰나	단점/장점
skill-forge	You already know what skill you want — skip the detection layer	No auto-detection
Manual SKILL.md authoring	One-off skill	No adaptive learning

gsd-skill-creator