cost-metering

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

<objective> Track Claude API costs across sessions with budget alerts, model routing optimization, and spend reporting. Integrates with workflow-orchestrator's cost gate for automated budget enforcement. </objective>

<quick_start> Check current spend:

bash

cat ~/.claude/daily-cost.json 2>/dev/null || echo "No tracking yet"

Initialize tracking:

bash

mkdir -p ~/.claude
echo '{"date":"'$(date +%Y-%m-%d)'","spent":0,"budget_monthly":100,"budget_daily":5}' > ~/.claude/daily-cost.json

</quick_start>

<success_criteria>

Daily cost tracking initialized at
```
~/.claude/daily-cost.json
```
Budget alerts fire at 50% (info), 80% (warn), and 95% (block) thresholds
Model routing applied: Haiku for search/classify, Sonnet for code, DeepSeek for bulk
Cost-per-feature metric available via portfolio-artifact integration
Monthly spend stays within configured budget ($100/mo default) </success_criteria>

<triggers> - "cost check", "budget status", "how much spent" - "optimize costs", "cost tracking", "budget alert" - "model routing", "cheaper model", "cost report" </triggers>

<objective> 跨会话跟踪Claude API成本，提供预算警报、模型路由优化以及支出报告。与workflow-orchestrator的成本网关集成，实现自动化预算管控。 </objective>

<quick_start> 查看当前支出：

bash

cat ~/.claude/daily-cost.json 2>/dev/null || echo "No tracking yet"

初始化跟踪：

bash

mkdir -p ~/.claude
echo '{"date":"'$(date +%Y-%m-%d)'","spent":0,"budget_monthly":100,"budget_daily":5}' > ~/.claude/daily-cost.json

</quick_start>

<success_criteria>

每日成本跟踪已初始化至
```
~/.claude/daily-cost.json
```
预算警报在50%（提示）、80%（警告）和95%（拦截）阈值触发
已应用模型路由：Haiku用于搜索/分类，Sonnet用于代码生成，DeepSeek用于批量处理
通过portfolio-artifact集成可获取按功能划分的成本指标
月度支出保持在配置的预算范围内（默认100美元/月） </success_criteria>

<triggers> - "成本核查"、"预算状态"、"支出金额" - "成本优化"、"成本跟踪"、"预算警报" - "模型路由"、"更经济的模型"、"成本报告" </triggers>

Model Rates (Current)

当前模型费率

Model	Input/1M tokens	Output/1M tokens	Typical Use
Claude Opus 4	$15.00	$75.00	Architecture, complex reasoning
Claude Sonnet 4.5	$3.00	$15.00	Code generation, standard tasks
Claude Haiku 4.5	$0.25	$1.25	Search, classification, simple
DeepSeek V3	$0.27	$1.10	Bulk processing
GROQ Llama 3.3 70B	$0.59	$0.79	Fast inference
Voyage Embeddings	$0.10	—	Embeddings

模型	每1M输入代币成本	每1M输出代币成本	典型用途
Claude Opus 4	$15.00	$75.00	架构设计、复杂推理
Claude Sonnet 4.5	$3.00	$15.00	代码生成、标准任务
Claude Haiku 4.5	$0.25	$1.25	搜索、分类、简单任务
DeepSeek V3	$0.27	$1.10	批量处理
GROQ Llama 3.3 70B	$0.59	$0.79	快速推理
Voyage Embeddings	$0.10	—	向量嵌入

Budget Configuration

预算配置

~/.claude/daily-cost.json

json

{
  "date": "2026-02-07",
  "spent": 2.40,
  "budget_monthly": 100,
  "budget_daily": 5,
  "alerts": {
    "info": 0.5,
    "warn": 0.8,
    "block": 0.95
  }
}

json

{
  "date": "2026-02-07",
  "spent": 2.40,
  "budget_monthly": 100,
  "budget_daily": 5,
  "alerts": {
    "info": 0.5,
    "warn": 0.8,
    "block": 0.95
  }
}

Alert Thresholds

警报阈值

Threshold	% Budget	Action
Info	50%	Display: "50% of monthly budget used"
Warn	80%	Yellow alert: "⚠️ 80% budget — consider model downgrade"
Block	95%	Red alert: "🛑 95% budget — require explicit override to continue"

阈值	预算占比	操作
提示	50%	显示："已使用50%月度预算"
警告	80%	黄色警报："⚠️ 已使用80%预算 — 考虑降级模型"
拦截	95%	红色警报："🛑 已使用95%预算 — 需要明确覆盖才能继续"

Cost Optimization Strategies

成本优化策略

1. Model Routing (biggest impact)

1. 模型路由（影响最大）

Task	Expensive	Optimized	Savings
File search	Sonnet ($3/1M)	Haiku ($0.25/1M)	92%
Code review	Sonnet ($3/1M)	Haiku ($0.25/1M)	92%
Classification	Sonnet ($3/1M)	Haiku ($0.25/1M)	92%
Bulk processing	Sonnet ($3/1M)	DeepSeek ($0.27/1M)	91%

Rule: If the task doesn't generate code, use Haiku. If it doesn't need Claude, use DeepSeek.

任务	高成本方案	优化后方案	成本节省比例
文件搜索	Sonnet（$3/1M）	Haiku（$0.25/1M）	92%
代码审查	Sonnet（$3/1M）	Haiku（$0.25/1M）	92%
分类任务	Sonnet（$3/1M）	Haiku（$0.25/1M）	92%
批量处理	Sonnet（$3/1M）	DeepSeek（$0.27/1M）	91%

规则： 如果任务不需要生成代码，使用Haiku。如果不需要Claude，使用DeepSeek。

2. Context Management

2. 上下文管理

Keep SKILL.md files under 200 lines (progressive disclosure)
Load reference files only when needed
Use
```
Explore
```
agent with
```
haiku
```
model for codebase search
Avoid reading entire files — use Grep to find specific lines

保持SKILL.md文件在200行以内（渐进式披露）
仅在需要时加载参考文件
使用
```
Explore
```
agent搭配
```
haiku
```
模型进行代码库搜索
避免读取整个文件 — 使用Grep查找特定行

3. Task Batching

3. 任务批处理

Group related searches into one Explore agent call
Use parallel subagents (haiku) instead of serial sonnet calls
Combine file reads when possible

将相关搜索合并为一次Explore agent调用
使用并行子agent（haiku）替代串行sonnet调用
尽可能合并文件读取操作

Tracking Commands

跟踪命令

Daily Spend Check

每日支出核查

bash

cat ~/.claude/daily-cost.json | jq '{date, spent, remaining: (.budget_daily - .spent), pct: ((.spent / .budget_monthly) * 100 | floor)}'

bash

cat ~/.claude/daily-cost.json | jq '{date, spent, remaining: (.budget_daily - .spent), pct: ((.spent / .budget_monthly) * 100 | floor)}'

Weekly Report

每周报告

bash

undefined

bash

undefined

Aggregate daily logs

汇总每日日志

cat ~/.claude/cost-log.jsonl | jq -s 'group_by(.phase) | map({phase: .[0].phase, total: (map(.est_cost) | add), count: length})'

undefined

cat ~/.claude/cost-log.jsonl | jq -s 'group_by(.phase) | map({phase: .[0].phase, total: (map(.est_cost) | add), count: length})'

undefined

Monthly Summary

月度汇总

bash

cat ~/.claude/cost-log.jsonl | jq -s '{
  total: (map(.est_cost) | add),
  by_model: (group_by(.model) | map({model: .[0].model, cost: (map(.est_cost) | add)})),
  by_phase: (group_by(.phase) | map({phase: .[0].phase, cost: (map(.est_cost) | add)}))
}'

bash

cat ~/.claude/cost-log.jsonl | jq -s '{
  total: (map(.est_cost) | add),
  by_model: (group_by(.model) | map({model: .[0].model, cost: (map(.est_cost) | add)})),
  by_phase: (group_by(.phase) | map({phase: .[0].phase, cost: (map(.est_cost) | add)}))
}'

Integration Points

集成点

System	How
workflow-orchestrator	Cost gate checks budget before workflows
subagent-teams	Model selection uses cost tiers
agent-capability-matrix	Includes model recommendations
portfolio-artifact	Reports cost-per-feature metrics
End Day protocol	Logs daily costs, updates MTD
TaskCreate/TaskUpdate	Zero API cost — local UI tool for progress tracking
TeamCreate/SendMessage	Zero API cost — local coordination (but spawned agents incur model costs)

系统	集成方式
workflow-orchestrator	工作流执行前通过成本网关检查预算
subagent-teams	模型选择采用成本分层策略
agent-capability-matrix	包含模型推荐内容
portfolio-artifact	报告按功能划分的成本指标
End Day协议	记录每日成本，更新月度累计数据
TaskCreate/TaskUpdate	零API成本 — 本地UI工具用于进度跟踪
TeamCreate/SendMessage	零API成本 — 本地协作工具（但生成的agent会产生模型成本）

Storage

存储结构

~/.claude/
├── daily-cost.json          # Current day's budget + spend
├── cost-log.jsonl           # Append-only operation log
└── portfolio/
    └── daily-metrics.jsonl  # Includes cost-per-feature

Deep dive: See

reference/cost-tracking-guide.md

reference/budget-templates.md

~/.claude/
├── daily-cost.json          # 当日预算与支出数据
├── cost-log.jsonl           # 追加式操作日志
└── portfolio/
    └── daily-metrics.jsonl  # 包含按功能划分的成本数据

深入了解： 请查看

reference/cost-tracking-guide.md

、

reference/budget-templates.md