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.
npx agentmods add commands/teamspwk/nova/claude-mdgit clone --depth 1 https://github.com/TeamSPWK/novaWhat 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 | $0.00035 | $0.00792 |
| Opus 5 | $0.00017 | $0.00396 |
| Sonnet 5 | $0.00007 | $0.00158 |
| Haiku 4.5 | $0.00003 | $0.00079 |
Grade A, and why
claude-md 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 2d 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
/nova:claude-md
프로젝트의 에이전트 지침 표면을 설계·정리한다. CLAUDE.md를 운영 헌법 + 라우터로 유지하고, 상세 절차·경로별 규칙·강제 가드·현재 상태를 올바른 위치로 분리한다.
적용 스킬
skills/claude-md/SKILL.md— 신규/기존 프로젝트 지침 감사, 분리 기준, 생성 템플릿
사용법
/nova:claude-md # 처음 사용자용 가이드 + 현재 레포 감사 시작
/nova:claude-md --check # 수정 없이 감사만
/nova:claude-md --apply # 승인된 재구성안 적용
/nova:claude-md --new # 신규 프로젝트 지침 골격 생성
/nova:claude-md --adopt # 기존 CLAUDE.md/AGENTS.md 재구성
/nova:claude-md --global-karpathy # 개인 전역 Karpathy 압축 원칙 추가/갱신 여부 확인
인자가 없으면 Guided Mode로 실행한다. 먼저 60초짜리 짧은 안내를 보여주고, 현재 레포에 CLAUDE.md/AGENTS.md가 있는지에 따라 --check 또는 --new에 해당하는 안전한 제안 흐름으로 이어간다.
실행 규칙
- 먼저
skills/claude-md/SKILL.md를 읽고 그 절차를 따른다. - 기존 지침 파일이 있으면 즉시 덮어쓰지 말고 섹션별
keep / move / enforce / local-only / remove분류표를 만든다. - 신규 프로젝트는 모르는 정보를 추측하지 말고
TODO(owner)또는 질문으로 남긴다. - hard guard는 현재 enforcement 상태(advisory / enforced / enforcement-optional)를 표시만 한다. 사용자가 명시적으로 요청하지 않으면
.claude/settings.json을 자동 생성·수정하지 않는다. 권한/차단 설정 적용은 사용자 결정 사항으로 안내한다. - 정리 결과에는 반드시 다음 에이전트가 같은 기준을 적용할 수 있는
Instruction Placement Contract를 남긴다. - 인자가 없으면 Guided Mode를 먼저 출력한다. 사용자가 플래그를 몰라도 다음 행동을 선택할 수 있게 한다.
--global-karpathy는 프로젝트 파일이 아니라~/.claude/CLAUDE.md,~/.codex/AGENTS.md만 대상으로 하며, 수정 전 반드시 yes/no 확인을 받는다.
Output
━━━ Agent Instructions Audit ━━━━━━━━━━━━━
Mode: Existing / New / Nested
Loaded: CLAUDE.md N lines, AGENTS.md N lines, rules N, skills N, settings yes/no
Verdict:
- Size: PASS/WARN/FAIL
- Duplication: PASS/WARN/FAIL
- Enforcement gaps: N
- Cross-agent bridge: PASS/WARN/FAIL
Placement Plan:
| Source section | Classification | Destination | Reason |
Proposed files:
| File | Action | Notes |
Questions:
- {추측하면 안 되는 정보}
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Input
$ARGUMENTS
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.
- 2d ago First seen · 64 lines · 35 tokens per session scan A db8247143c4c
claude-md is a command published in the GitHub repository TeamSPWK/nova (2 stars, last pushed 3d ago), licensed MIT. It adds 35 tokens to every session and 792 once invoked, about $0.0002 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
loopx
Run the loopx setup helper and read its output.
grow-scanner
Autonomous task: increase real-world adoption of the AgentDiscover scanner.
gedd-chat
You are a GEDD coaching assistant. You guide the user through building a golden evaluation dataset for their AI agent using Open Coding methodology, then help them evaluate and annotate responses — all without leaving Claude Code.
gedd
You are a GEDD coaching agent. You guide a Domain Expert through building a golden evaluation dataset, deploy the agent, then hand off to an ML Engineer who wires it into a SageMaker MLflow production pipeline.
gedd-status
Read session.json (use the Read tool) and display a concise dashboard. If the file doesn't exist, say so and suggest running /gedd to start.
confirm-revisions
Toggle Second Pass re-judge after revision. true = verify lift in a loop (more tokens, more confidence). false = trust judge's feedback, ship after one revision (default, cheaper).