Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/sodam-ai/SoDam-Harness-EngWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/sodam-ai/sodam-harness-eng/install)<a href="https://agentmods.dev/commands/sodam-ai/sodam-harness-eng/install"><img src="https://agentmods.dev/badge/commands/sodam-ai/sodam-harness-eng/install/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/sodam-ai/sodam-harness-eng/install"><img src="https://agentmods.dev/badge/commands/sodam-ai/sodam-harness-eng/install.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00015 | $0.00407 |
| Opus 5 | $0.00008 | $0.00204 |
| Sonnet 5 | $0.00003 | $0.00081 |
| Haiku 4.5 | $0.00002 | $0.00041 |
Grade A, and why
install scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
사용자가 방금 SoDamHarness를 설치했거나 처음 시작합니다. 아래를 쉬운 한국어로 안내하세요.
- 설치 확인 인사: "안전벨트(SoDamHarness)가 켜졌어요 ✅" 라고 알려줍니다.
- 사전 점검(Node.js):
node --version을 실행해 Node.js가 있는지 확인합니다.- 있으면: 버전을 알려주고 "준비 끝났어요" 라고 합니다.
- 없으면: "안전 기능이 동작하려면 Node.js가 필요해요. https://nodejs.org 에서 LTS 버전을 설치한 뒤 Claude Code를 다시 켜 주세요." 라고 쉽게 안내합니다.
- 무엇이 켜졌는지 한 줄 설명: "이제 파일을 지우거나 덮어쓰는 위험한 작업을 하기 전에 자동으로 멈추고 물어보고, 먼저 백업해요. 잘못되면
/sodam-harness:undo로 되돌릴 수 있어요." - 지금 할 것 하나: "한번 해볼까요? 작은 연습 폴더에서 AI에게 자연어로 시켜 보세요. 예) '여기에 간단한 소개 페이지 만들어 줘'."
- 백업 위치 안내: 백업은 사용자 홈 폴더의
.sodamharness/backups/에 저장된다고 절대 경로로 알려줍니다(예:C:\Users\<이름>\.sodamharness\backups\).
과장 금지: "100% 안전"이라고 말하지 마세요. 정해진 위험을 막아주지만 모든 경우를 막지는 못한다고 정직하게 덧붙이세요.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 16 lines · 15 tokens per session scan A b8ba6a376c77
install is a command published in the GitHub repository sodam-ai/SoDam-Harness-Eng (1 stars, last pushed today), licensed Apache-2.0. It adds 15 tokens to every session and 407 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
start-issue
You are beginning work on GitHub issue $ARGUMENTS.
hallucination-audit
Detect speculation-as-diagnosis, invented causality, and unverified claims in text. Use after producing an answer or reviewing sub-agent output.
start
A Korean-language start command for beginning a repair loop after refining a goal, making a backup, giving guidance, and confirming the plan.
uninstall
Instructions for removing a plug-in and handling the local data it leaves behind, including state and backups. The instructions are written in Korean.
checkup
A Korean-language command for scheduled checks that only reports what it finds. It examines previously saved goals but never changes or repairs them.
help
A numbered help menu explaining what Sodamloop can do.