---
Analyzes command complexity and recommends cost-optimal AI models for execution.
npx claudepluginhub lofibrainwav/hyodo$ARGUMENTS 작업의 예상 비용을 CostAwareRouter로 분석합니다.
cost_tiers:
FREE:
model: qwen3:8b (Ollama)
provider: ollama
cost_per_1k_tokens: $0.000
quality_score: 0.70
max_tokens: 8192
best_for:
- 단순 조회 (read, list, show)
- 도움말 요청 (help, explain)
- 검색 (search, find)
CHEAP:
model: claude-haiku-4-5-20251001
provider: anthropic
cost_per_1k_tokens: $0.00025
quality_score: 0.85
max_tokens: 4096
best_for:
- 일반 구현 (implement, create, add)
- 디버깅 (debug, fix)
- 테스트 작성 (test)
- 업데이트 (update, modify)
EXPENSIVE:
model: claude-opus-4-5-20251101
provider: anthropic
cost_per_1k_tokens: $0.015
quality_score: 0.98
max_tokens: 8192
best_for:
- 프로덕션 배포 (production, deploy)
- 보안 관련 (auth, secret, password)
- 아키텍처 결정 (refactor, architect)
- 삭제 작업 (delete, drop, destroy)
- 마이그레이션 (migration)
production, deploy, delete, dropauth, secret, password, credentialmigration, refactor, architectimplement, create, add, updatemodify, test, debug, fixread, list, show, explainsearch, find, help, docsrequires_approval: +2점dry_run: -1점cost_estimate:
command: "$ARGUMENTS"
analysis:
complexity_score: [점수]
matched_keywords:
high: [고복잡도 키워드들]
medium: [중복잡도 키워드들]
low: [저복잡도 키워드들]
recommendation:
tier: [FREE | CHEAP | EXPENSIVE]
model: [모델명]
provider: [ollama | anthropic]
quality_score: [0.70 | 0.85 | 0.98]
cost_projection:
estimated_tokens: [예상 토큰 수]
cost_per_1k: $[비용]
total_estimated: $[총 예상 비용]
strategist_note:
jang_yeong_sil: "[기술적 복잡도 평가]"
yi_sun_sin: "[리스크 평가]"
shin_saimdang: "[사용자 경험 평가]"
# 단순 조회 → FREE
/cost-estimate "list all files in src directory"
# 결과: FREE, qwen3:8b, $0.00
# 일반 구현 → CHEAP
/cost-estimate "implement a new utility function for date formatting"
# 결과: CHEAP, claude-haiku-4-5, ~$0.0005
# 프로덕션 배포 → EXPENSIVE
/cost-estimate "deploy authentication module to production"
# 결과: EXPENSIVE, claude-opus-4-5, ~$0.03
"백성의 부담을 최소화하라"
비용 최적화는 善의 원칙입니다:
packages/afo-core/api/chancellor_v2/orchestrator/cost_aware_router.py