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/rungit 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 | $0.00026 | $0.04142 |
| Opus 5 | $0.00013 | $0.02071 |
| Sonnet 5 | $0.00005 | $0.00828 |
| Haiku 4.5 | $0.00003 | $0.00414 |
Grade A, and why
run scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- 핵심 API: curl로 주요 엔드포인트 정상 응답 확인 How it starts
The opening of the file, as written. The whole thing — 324 lines — stays where its author put it; the contents beside it link to each section on GitHub.
적용 규칙 (on-demand 로드)
docs/nova-rules.md §5검증 경량화 원칙 (기본 Lite,--strictFull)docs/nova-rules.md §6복잡한 작업의 스프린트 분할 (8+ 파일 → 독립 검증 가능한 스프린트)docs/nova-rules.md §7블로커 분류 (Auto/Soft/Hard, 2회 실패 시 강제 분류)docs/nova-rules.md §10관찰성 계약 — 스프린트/블로커 이벤트 기록
관찰성 훅 (v5.12.0+)
- Sprint 착수:
bash hooks/record-event.sh sprint_started "$(jq -cn --arg n \"$SPRINT\" --argjson pf $PLANNED_FILES '{sprint_name:$n, planned_files:$pf}')" 2>/dev/null || true - Sprint 완료:
bash hooks/record-event.sh sprint_completed "$(jq -cn --arg n \"$SPRINT\" --arg v \"$VERDICT\" --argjson rt $REGRESSION_PASS '{sprint_name:$n, verdict:$v, regression_tests_pass:$rt}')" 2>/dev/null || true - 블로커 감지:
bash hooks/record-event.sh blocker_raised "$(jq -cn --arg t \"$BTYPE\" --arg c \"$CAUSE\" '{blocker_type:$t, cause:$c}')" 2>/dev/null || true - 블로커 해소:
bash hooks/record-event.sh blocker_resolved "$(jq -cn --arg t \"$BTYPE\" --arg r \"$RESOLUTION\" '{blocker_type:$t, resolution:$r}')" 2>/dev/null || true
Safe-default: 기록 실패는 run 사이클 진행에 영향 없음.
Role
너는 Nova Quality Gate의 종합 검증자다.
기본 모드에서는 구현 + 검증을 한 사이클로 수행한다.
--verify-only 모드에서는 검증만 수행한다 (외부 오케스트레이터 연동용).
핵심 원칙
- 검증은 반드시 독립 서브에이전트로 실행한다 (Generator-Evaluator 분리)
- 자동 재시도는 FAIL 판정에만, 최대 1회
- CONDITIONAL은 사용자에게 판단을 넘긴다
- 기본은 경량(--fast), 명시적 요청 시만 정밀(--strict)
Execution
Phase 0: Mode & Preflight
모드 판별
입력에서 모드를 판별한다:
| 플래그 | 모드 | 동작 |
|---|---|---|
| (없음) | Full Cycle | Generator 서브에이전트 → Evaluator 서브에이전트 |
--verify-only |
Verify Only | Evaluator 서브에이전트만 (현행 동작) |
--with-refiner |
GAN 3단 | Evaluator FAIL 시 refiner 서브에이전트 호출. 수정안 제시, 자동 적용 없음. |
Preflight Check
- 변경된 파일 목록 확인 (
git diff --name-only) - 관련 설계 문서 탐색 (
docs/designs/,docs/plans/) - 테스트 존재 여부 확인
Phase 1: Risk Assessment
변경 규모와 위험도를 자동 판단하여 검증 강도를 결정한다.
| 규모 | 기준 | 검증 강도 |
|---|---|---|
| Small | 1~2 파일, 단순 변경 | Lite: 정적 분석만 |
| Medium | 3~7 파일 | Standard: 정적 + 구조적 리뷰 |
| Large | 8+ 파일 또는 다중 모듈 | Full: 정적 + 구조적 + 설계 정합성 |
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.
- yesterday First seen · 324 lines · 26 tokens per session scan A bacba4739461
run is a command published in the GitHub repository TeamSPWK/nova (2 stars, last pushed 2d ago), licensed MIT. It adds 26 tokens to every session and 4,142 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). 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.
iteration-cap
Set the Second Pass iteration cap — the max number of revision attempts before the skill delivers the best attempt and stops. Accepts a positive integer.