A multi-perspective consultation command that asks Claude, GPT, and Gemini for views on a question and analyzes how much they agree. It is intended for choices without one obvious correct answer.

Use it to compare options for technology stacks, architectures, and designs; it can cross-check answers when the required API keys are available or fall back to agent-based responses.

It gives you several independent opinions for architecture, technology, and design trade-offs instead of relying on one viewpoint. It is less suitable for factual checks, syntax errors, or simple API questions.

ask — Command by TeamSPWK

Install

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.

agentmods

npx agentmods add commands/teamspwk/nova/ask

Clone the repo

git clone --depth 1 https://github.com/TeamSPWK/nova

Made for: Claude Code.

Per session 33 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 2,142 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00033	$0.02142
Opus 5	$0.00016	$0.01071
Sonnet 5	$0.00007	$0.00428
Haiku 4.5	$0.00003	$0.00214

Measured yesterday against content hash 9f189581b6ae, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

ask 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 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.

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.

.claude/commands/ask.md · 205 lines

How it starts

The opening of the file, as written. The whole thing — 205 lines — stays where its author put it; the contents beside it link to each section on GitHub.

멀티 AI 다관점 자문을 실행한다. Claude + GPT + Gemini 3개 AI에게 동시에 질의하고 합의 수준을 분석한다.

적용 규칙 (on-demand 로드)

docs/nova-rules.md §10 관찰성 계약 — 합의 분석 직후 jury_verdict 이벤트 기록

관찰성 훅 (v5.12.0+)

합의 수준 판정 직후 반드시:

bash hooks/record-event.sh jury_verdict "$(jq -cn \
  --arg cl "$CONSENSUS_LEVEL" \
  --argjson cd "$CHANGED_DIRECTION" \
  '{consensus_level:$cl, changed_direction:$cd}')" 2>/dev/null || true

CONSENSUS_LEVEL: strong / partial / divergent (소문자)
CHANGED_DIRECTION: 자문 결과가 초기 방향을 뒤집었으면 true, 유지면 false

Safe-default: 기록 실패는 ask 결과 반환에 영향 없음.

Role

너는 Nova Engineering의 다관점 자문(Consult) 실행자다. 사용자의 질문을 여러 관점에서 수집하고, 결과를 종합하여 합의 분석을 제공한다.

적용 범위

적합: 기술 스택 선택, 아키텍처 판단, 설계 트레이드오프 등 "정답이 없는 판단"
부적합: 사실 확인(factual), 구문 오류, 단순 API 사용법 등 — 대부분 같은 답이 나오므로 합의가 높아도 무의미

사전 요구사항 (Mode A)

프로젝트 루트 .env에 다음 API 키를 설정한다 (최소 1개):

ANTHROPIC_API_KEY=sk-ant-...
OPENAI_API_KEY=sk-...
GEMINI_API_KEY=AI...

키 보유	동작
3개	교차검증 + 합의 분석 (최적)
2개	교차검증 (합의 분석은 GEMINI_API_KEY 필요)
1개	단일 AI 응답 + 에이전트 비교
0개	Mode B (에이전트 폴백) 자동 전환

선택적 모델 오버라이드 (.env):

CLAUDE_MODEL=claude-sonnet-4-6        # 기본 Claude 모델
CLAUDE_MODEL_OPUS=claude-opus-4-7     # --model opus 시
CLAUDE_MODEL_HAIKU=claude-haiku-4-5-20251001
OPENAI_MODEL=gpt-5.4                  # 기본 GPT 모델
GEMINI_MODEL=gemini-3-flash-preview   # 기본 Gemini 모델

Execution

Mode A: API 다관점 수집 (.env에 API 키가 1개 이상 있을 때)

권장: API 키가 있으면 항상 Mode A를 사용한다. Mode A만이 진짜 독립적인 멀티 AI 교차검증이다.

인자가 없으면 사용자에게 질문을 받는다.
MCP 도구 x_verify를 호출한다:
- question: 질문 내용
- no_save: --no-save 옵션 시 true
- selected_ais: --claude, --gpt, --gemini 옵션에 해당하는 AI 배열 (미지정 시 생략)
- claude_model: --model opus|sonnet|haiku 옵션 (미지정 시 생략)
실행 결과를 사용자에게 보여준다.

Mode B: 에이전트 다관점 수집 (API 키가 없거나 --agent 옵션 사용 시)

API 키 없이도 다관점 수집을 수행한다. 3개 병렬 에이전트를 서로 다른 전문가 관점으로 실행한다.

⚠️ 한계: Mode B는 교차검증이 아닌 다관점 시뮬레이션입니다. 동일 모델의 편향이 반복됩니다. 진짜 독립 검증이 필요하면 반드시 Mode A를 사용하세요.

Read the full file on GitHub · 205 lines

Changes

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 · 205 lines · 33 tokens per session scan A 9f189581b6ae

Subscribe to this mod's changes

ask is a command published in the GitHub repository TeamSPWK/nova (2 stars, last pushed yesterday), licensed MIT. It adds 33 tokens to every session and 2,142 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.

Backend and APIs Code review agent-ops ai-agents ai-quality anthropic claude-code claude-code-plugin

ask

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

적용 규칙 (on-demand 로드)

관찰성 훅 (v5.12.0+)

Role

적용 범위

사전 요구사항 (Mode A)

Execution

Mode A: API 다관점 수집 (.env에 API 키가 1개 이상 있을 때)

Mode B: 에이전트 다관점 수집 (API 키가 없거나 --agent 옵션 사용 시)

What this file has done since we first saw it

Other commands, from other repositories

loopx

grow-scanner

gedd-chat

gedd

gedd-status

iteration-cap