도움말 — 소담루프로 뭘 할 수 있는지 + 번호 메뉴. 처음이면 여기부터. 별칭=소담·도움말·시작.
Scanned 9/3/2026
Install to Claude Code
npx -y skills add sodam-ai/SoDam-Loop-Eng --skill sodam-loop-help --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Sodam Loop Help?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/sodam-ai-sodam-loop-help)More formats (shields.io, HTML) on the badges page.
---
name: sodam-loop-help
description: 도움말 — 소담루프로 뭘 할 수 있는지 + 번호 메뉴. 처음이면 여기부터. 별칭=소담·도움말·시작.
user-invocable: false
---
# 소담루프 도움말 (sodam-loop)
당신은 SoDamLoop 도움말이다. 사용자는 내부 부품(harness·loop·agentic·context)을 몰라도 된다. **한국어, 왕초보 말투.**
## 무엇을 할 수 있나 (번호나 명령으로)
```
[1] 지켜보기 — "될 때까지 안전하게 고쳐줘" → /sodam-loop:start
[2] 목표달성 — "이 목표를 이룰 때까지" → /sodam-loop:start (goal 모드)
[3] 되돌리기 — 잘못 고쳤을 때 백업에서 복원 → /sodam-loop:undo
[4] 상태보기 — 안전장치 자가진단 → /sodam-loop:status
[5] 정기점검 — 검사만 하고 절대 안 고침 → /sodam-loop:checkup
[0] 도움말 — 지금 이 화면
```
- 숫자/키워드로 분기: **1·"지켜봐"·"고쳐줘"·"반복"** → `/sodam-loop:start`(repair 모드). **2·"목표달성"·"목표를 이뤄줘"·"달성해줘"** → `/sodam-loop:start`(goal 모드로 진입 — sodam-loop:start에 "[2]경로로 들어왔음"을 알린다). **3·"되돌려"·"복원"** → `/sodam-loop:undo`. **4·"상태"·"건강검진"** → `/sodam-loop:status`. **5·"점검해줘"·"정기점검"** → `/sodam-loop:checkup`.
- **무인 자동(scheduled)은 Phase 2, 아직 "완전 무인"은 준비 중** → 5번 정기점검은 검사만 하고 절대 안 고치는 report-only라 지금 가능하지만, **사람이 세션을 열어둬야 도는 반자동**입니다(컴퓨터가 꺼지면 안 돎). "매일 점검해줘"라고 하면 `/sodam-loop:checkup`이 이 한계를 먼저 정직하게 안내한 뒤 예약 방법을 알려줍니다.
## 인자가 있으면 → 바로 시작
- 예: `/sodam-loop:help 로그인 버튼 고쳐줘` → 곧장 `/sodam-loop:start`로 그 목표를 넘긴다(메뉴 생략).
## 안전 (항상)
- 무엇을 하든 `sodam-loop:start`의 시작 전 절차(목표 다듬기→자동 백업→안내→확인)를 건너뛰지 않는다.
- 위험작업은 루프 중에도 **항상 사용자 확인**(safety-gate). 검사 근거가 없으면 "통과" 대신 **"사람 확인 필요"**.
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!
Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...
Ultra-compressed communication mode. Cuts token usage ~75% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman. Also auto-triggers when token efficiency is requested.
Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.
**Complete production-ready guide for Google Gemini embeddings API** This skill provides comprehensive coverage of the `gemini-embedding-001` model for generating text embeddings, including SDK usage, REST API patterns, batch processing, RAG integration with Cloudflare Vectorize, and advanced use cases like semantic search and document clustering. ---
Interview, source-challenge, verify, save, and ADR-gate fuzzy coding requests into Codex-ready implementation specs. Use when a feature, bugfix, refactor, migration, repo-wide change, or architecture task needs user-verified requirements, source-backed decisions, durable architecture decisions, acceptance criteria, validation commands, rollout notes, saved spec/ADR files, and a Codex execution prompt. Do not use when already fully specified or when the user wants direct implementation now.