使用AWS CUR V2账单分析Kiro用户的Credit用量

一、背景

由于目前（2026-01）Kiro的控制台Dashboard暂时不能显示单个用户的Credit使用情况，因此建议通过计费系统进行查询。本文讲述如何使用Athena查询CUR V2获得Kiro用户的Credit使用情况。

1、Kiro计费模式

Kiro目前提供三个级别的订阅，分别是：

Pro: 内含 1000 Credit；
Pro+：内含 2000 Credit；
Power：内含 10000 Credit；

以上，超出套餐内的使用量之后，额外消耗的Credit按照$0.04/credit计费。

官方对Credit的解释如下：https://kiro.dev/pricing/#what-is-a-credit

Credit（积分）是响应用户提示的工作单位。简单的提示可能消耗不到 1 个积分。
更复杂的提示，例如执行规格任务，通常会消耗超过 1 个积分。
此外，不同的模型以不同的速率消耗积分，使用Sonnet 4 执行提示比使用 Auto 执行消耗更多积分。

由于目前（2026-01）Kiro的控制台Dashboard暂时不能显示单个用户的Credit使用情况，因此建议通过AWS云的账单系统的CUR数据进行查询。

2、CUR V2介绍

AWS CUR V2（Cost and Usage Report 2.0）是 AWS 在 2023 年 re:Invent 发布的新版成本和用量报告，为用户给出完整的账单计费原始数据并保存到S3存储桶中。CUR V2支持csv（gzip压缩）或者parquet格式，可以使用Athena服务的SQL语句进行查询。

二、创建CUR V2

1、创建S3存储桶

创建一个用于保存CUR数据的存储桶，放在美东1(us-east-1)区域。存储桶设置全默认即可。

2、开启CUR V2

进入AWS控制台，进入Billing and Cost Management服务，在左侧菜单下，找到菜单项Cost and Usage Analysis，再从中找到Data Exports，点击进去。

点击Create创建按钮，选择如下：

类型选择Standard data export
名字自行拟定
Data table content settings选择CUR 2.0
Billing view设置不用修改
Additional export content设置不用修改
Time granularity选择为daily或Hourly均可
SQL statement默认已经选中了所有列，不用修改
Data export delivery options的Compression type and file format选择Parquet - Parquet
Data export storage settings位置输入上一步创建的S3存储桶和路径，建议在存储桶根目录下增加子目录（prefix），例如s3://cur-v2-myaws/cur-export，有子目录便于未来创建Athena表格

其他设置无须修改，点击创建。

3、等待24-48小时生成第一批CUR数据

注意：

CUR有较大延迟，通常在1天以上，创建CUR时候有个时间颗粒度选项，hourly或者daily，这表示CUR明细的颗粒度，但不意味着选hourly就是每小时推送CUR数据。CUR数据是按天推送的
CUR只记录打开一刻起的数据。例如本月15日启用了CUR，那么17日左右开始生成15-16日的数据，15日之前的将不会提供CUR

三、使用Athena查询CUR数据获取Kiro使用量

在等待24-48小时后，CUR有了数据，继续创建表格做查询。

1、使用Athena创建表

进入Athena服务，在默认database下执行如下SQL创建表。注意替换以下SQL中的S3存储桶的路径’s3://cur-lxy-us-east-1/cur-v2/cur-v2/data/‘为实际的CUR保存目录下自动生成的data目录。你可以从S3存储桶服务界面去查看目录，直到发现最后一层data目录。

CREATE EXTERNAL TABLE `cur-v2data`(
  `bill_bill_type` string, 
  `bill_billing_entity` string, 
  `bill_billing_period_end_date` timestamp, 
  `bill_billing_period_start_date` timestamp, 
  `bill_invoice_id` string, 
  `bill_invoicing_entity` string, 
  `bill_payer_account_id` string, 
  `bill_payer_account_name` string, 
  `cost_category` map<string,string>, 
  `discount` map<string,double>, 
  `discount_bundled_discount` double, 
  `discount_total_discount` double, 
  `identity_line_item_id` string, 
  `identity_time_interval` string, 
  `line_item_availability_zone` string, 
  `line_item_blended_cost` double, 
  `line_item_blended_rate` string, 
  `line_item_currency_code` string, 
  `line_item_legal_entity` string, 
  `line_item_line_item_description` string, 
  `line_item_line_item_type` string, 
  `line_item_net_unblended_cost` double, 
  `line_item_net_unblended_rate` string, 
  `line_item_normalization_factor` double, 
  `line_item_normalized_usage_amount` double, 
  `line_item_operation` string, 
  `line_item_product_code` string, 
  `line_item_resource_id` string, 
  `line_item_tax_type` string, 
  `line_item_unblended_cost` double, 
  `line_item_unblended_rate` string, 
  `line_item_usage_account_id` string, 
  `line_item_usage_account_name` string, 
  `line_item_usage_amount` double, 
  `line_item_usage_end_date` timestamp, 
  `line_item_usage_start_date` timestamp, 
  `line_item_usage_type` string, 
  `pricing_currency` string, 
  `pricing_lease_contract_length` string, 
  `pricing_offering_class` string, 
  `pricing_public_on_demand_cost` double, 
  `pricing_public_on_demand_rate` string, 
  `pricing_purchase_option` string, 
  `pricing_rate_code` string, 
  `pricing_rate_id` string, 
  `pricing_term` string, 
  `pricing_unit` string, 
  `product` map<string,string>, 
  `product_comment` string, 
  `product_fee_code` string, 
  `product_fee_description` string, 
  `product_from_location` string, 
  `product_from_location_type` string, 
  `product_from_region_code` string, 
  `product_instance_family` string, 
  `product_instance_type` string, 
  `product_instancesku` string, 
  `product_location` string, 
  `product_location_type` string, 
  `product_operation` string, 
  `product_pricing_unit` string, 
  `product_product_family` string, 
  `product_region_code` string, 
  `product_servicecode` string, 
  `product_sku` string, 
  `product_to_location` string, 
  `product_to_location_type` string, 
  `product_to_region_code` string, 
  `product_usagetype` string, 
  `reservation_amortized_upfront_cost_for_usage` double, 
  `reservation_amortized_upfront_fee_for_billing_period` double, 
  `reservation_availability_zone` string, 
  `reservation_effective_cost` double, 
  `reservation_end_time` string, 
  `reservation_modification_status` string, 
  `reservation_net_amortized_upfront_cost_for_usage` double, 
  `reservation_net_amortized_upfront_fee_for_billing_period` double, 
  `reservation_net_effective_cost` double, 
  `reservation_net_recurring_fee_for_usage` double, 
  `reservation_net_unused_amortized_upfront_fee_for_billing_period` double, 
  `reservation_net_unused_recurring_fee` double, 
  `reservation_net_upfront_value` double, 
  `reservation_normalized_units_per_reservation` string, 
  `reservation_number_of_reservations` string, 
  `reservation_recurring_fee_for_usage` double, 
  `reservation_reservation_a_r_n` string, 
  `reservation_start_time` string, 
  `reservation_subscription_id` string, 
  `reservation_total_reserved_normalized_units` string, 
  `reservation_total_reserved_units` string, 
  `reservation_units_per_reservation` string, 
  `reservation_unused_amortized_upfront_fee_for_billing_period` double, 
  `reservation_unused_normalized_unit_quantity` double, 
  `reservation_unused_quantity` double, 
  `reservation_unused_recurring_fee` double, 
  `reservation_upfront_value` double, 
  `resource_tags` map<string,string>, 
  `savings_plan_amortized_upfront_commitment_for_billing_period` double, 
  `savings_plan_end_time` string, 
  `savings_plan_instance_type_family` string, 
  `savings_plan_net_amortized_upfront_commitment_for_billing_period` double, 
  `savings_plan_net_recurring_commitment_for_billing_period` double, 
  `savings_plan_net_savings_plan_effective_cost` double, 
  `savings_plan_offering_type` string, 
  `savings_plan_payment_option` string, 
  `savings_plan_purchase_term` string, 
  `savings_plan_recurring_commitment_for_billing_period` double, 
  `savings_plan_region` string, 
  `savings_plan_savings_plan_a_r_n` string, 
  `savings_plan_savings_plan_effective_cost` double, 
  `savings_plan_savings_plan_rate` double, 
  `savings_plan_start_time` string, 
  `savings_plan_total_commitment_to_date` double, 
  `savings_plan_used_commitment` double, 
  `split_line_item_actual_usage` double, 
  `split_line_item_net_split_cost` double, 
  `split_line_item_net_unused_cost` double, 
  `split_line_item_parent_resource_id` string, 
  `split_line_item_public_on_demand_split_cost` double, 
  `split_line_item_public_on_demand_unused_cost` double, 
  `split_line_item_reserved_usage` double, 
  `split_line_item_split_cost` double, 
  `split_line_item_split_usage` double, 
  `split_line_item_split_usage_ratio` double, 
  `split_line_item_unused_cost` double)
PARTITIONED BY ( 
  `billing_period` string)
ROW FORMAT SERDE 
  'org.apache.hadoop.hive.ql.io.parquet.serde.ParquetHiveSerDe' 
STORED AS INPUTFORMAT 
  'org.apache.hadoop.hive.ql.io.parquet.MapredParquetInputFormat' 
OUTPUTFORMAT 
  'org.apache.hadoop.hive.ql.io.parquet.MapredParquetOutputFormat'
LOCATION
  's3://cur-lxy-us-east-1/cur-v2/cur-v2/data/'
TBLPROPERTIES (
  'classification'='parquet', 
  'compressionType'='none', 
  'partition_filtering.enabled'='true', 
  'typeOfData'='file')

随后即可在Athena默认数据库下，创建了名为cur-v2data的表。

2、查询使用量

(1) 查询本月所有和Kiro相关的计费条目（含Credit使用和订阅）

在Athena上运行如下SQL

-- 查询本月所有和Kiro相关的数据
SELECT 
    identity_time_interval as time_range,
    line_item_operation as cost_type,
    product_usagetype as usage_type,
    line_item_line_item_description as description,
    product_sku as sku,
    line_item_usage_amount as usage_amount,
    line_item_resource_id as user_id,
    line_item_blended_cost as blended_cost,
    line_item_unblended_cost as unblended_cost
FROM "default"."cur-v2data"
WHERE
    billing_period = '2026-01'
    AND product_location = 'US East (N. Virginia)'
    AND line_item_product_code = 'Kiro'
Order by time_range, usage_type
LIMIT 20;

注意：以上SQL查询结果中，user_id是对应的Kiro订阅时候的IAM Identity Center中的user的id。这个id不是用户名也不是邮箱，下文可通过关联查询获取用户真实姓名或邮箱。

(2) 查询本月内所有用户的Kiro Credit使用累计

WITH user_plans AS (
    SELECT DISTINCT
        SPLIT_PART(line_item_resource_id, 'user/', 2) as user_id,
        CASE
            WHEN product_usagetype LIKE '%Pro+%' THEN 'Pro+'
            WHEN product_usagetype LIKE '%Power%' THEN 'Power'
            WHEN product_usagetype LIKE '%Pro%' THEN 'Pro'
        END as subscription_plan
    FROM "default"."cur-v2data"
    WHERE
        billing_period = '2026-01'
        AND line_item_product_code = 'Kiro'
        AND line_item_operation = 'monthly-subscription'
)
SELECT
    p.subscription_plan,
    CASE
        WHEN p.subscription_plan = 'Pro' THEN 1000
        WHEN p.subscription_plan = 'Pro+' THEN 2000
        WHEN p.subscription_plan = 'Power' THEN 10000
    END as plan_credits,
    p.user_id,
    SUM(c.line_item_usage_amount) as total_credits
FROM "default"."cur-v2data" c
JOIN user_plans p ON SPLIT_PART(c.line_item_resource_id, 'user/', 2) = p.user_id
WHERE
    c.billing_period = '2026-01'
    AND c.line_item_product_code = 'Kiro'
    AND c.line_item_operation = 'Credits'
GROUP BY
    p.subscription_plan,
    p.user_id
ORDER BY
    CASE
        WHEN p.subscription_plan = 'Pro' THEN 1
        WHEN p.subscription_plan = 'Pro+' THEN 2
        WHEN p.subscription_plan = 'Power' THEN 3
    END,
    p.user_id

返回数据格式如下：

#	subscription_plan	plan_credits	user_id	total_credits
1	Pro	1000	843884c8-00c1-7046-e4e8-7ba5acc6cfc7	1.7210808227999999

(3) 查询哪些用户超过了订阅的Plan的总和（如果没有超额则显示为空）

WITH user_plans AS (
    SELECT DISTINCT
        SPLIT_PART(line_item_resource_id, 'user/', 2) as user_id,
        CASE
            WHEN product_usagetype LIKE '%Pro+%' THEN 'Pro+'
            WHEN product_usagetype LIKE '%Power%' THEN 'Power'
            WHEN product_usagetype LIKE '%Pro%' THEN 'Pro'
        END as subscription_plan
    FROM "default"."cur-v2data"
    WHERE
        billing_period = '2026-01'
        AND line_item_product_code = 'Kiro'
        AND line_item_operation = 'monthly-subscription'
),
user_usage AS (
    SELECT
        p.subscription_plan,
        CASE
            WHEN p.subscription_plan = 'Pro' THEN 1000
            WHEN p.subscription_plan = 'Pro+' THEN 2000
            WHEN p.subscription_plan = 'Power' THEN 10000
        END as plan_credits,
        p.user_id,
        SUM(c.line_item_usage_amount) as total_credits
    FROM "default"."cur-v2data" c
    JOIN user_plans p ON SPLIT_PART(c.line_item_resource_id, 'user/', 2) = p.user_id
    WHERE
        c.billing_period = '2026-01'
        AND c.line_item_product_code = 'Kiro'
        AND c.line_item_operation = 'Credits'
    GROUP BY
        p.subscription_plan,
        p.user_id
)
SELECT
    subscription_plan,
    plan_credits,
    user_id,
    total_credits,
    total_credits - plan_credits as overage
FROM user_usage
WHERE total_credits > plan_credits
ORDER BY
    CASE
        WHEN subscription_plan = 'Pro' THEN 1
        WHEN subscription_plan = 'Pro+' THEN 2
        WHEN subscription_plan = 'Power' THEN 3
    END,
    user_id

如果有用户超额，则显示用户清单，如果没有超额，则显示为空。

四、建立Kiro的用户名和用户ID的映射关系

本文采用从IAM Identity Center手工导出一个最新用户清单的办法，另外新建一个Athena表，然后做联合查询完成userid和用户名的对应关系。如果需要的话，稍微修改下，即可完成用户邮箱和用户id的映射。

1、使用AWSCLI查询IAM Identity Center上的用户ID映射关系

在配置了AWSCLI的环境上，使用如下命令，查询AWS当前使用的SSO的Instance ID。

aws sso-admin list-instances --region us-east-1

返回结果如下：

{
    "Instances": [
        {
            "InstanceArn": "arn:aws:sso:::instance/ssoins-72236887c48a0b06",
            "IdentityStoreId": "d-9066195b8c",
            "OwnerAccountId": "133129065110",
            "CreatedDate": "2025-12-22T11:36:00.192000+08:00",
            "Status": "ACTIVE"
        }
    ]
}

以上信息中的IdentityStoreId就是我们需要的信息。将这个ID复制下来，构建如下AWSCLI命令，导出用户清单。

aws identitystore list-users \
  --identity-store-id d-9066195b8c \
  --region us-east-1 \
  --query 'Users[*].{Username:UserName,UserId:UserId}' \
  --output json | jq -c '.[]' > user-list.jsonl

这个命令会将当前IAM Identity Center上的所有用户输出到一个JSON List文件。如果用户数量较多，AWSCLI会自动分批获取，时间会比较长。

生成的jsonl文件格式如下。

{"Username":"user01","UserId":"843884c8-00c1-7046-e4e8-7ba5acc6cfc7"}
{"Username":"user02","UserId":"04d86408-4001-7047-01d1-077885ab54fa"}

现在在执行AWSCLI的本地环境中获得了user-list.jsonl文件。接下来将这个文件导入到Athena的表中方便SQL查询。

2、在Athena内创建用户名和用户ID映射表

将以上文件上传到S3存储桶，例如路径是s3://cur-lxy-us-east-1/user-list/。然后登陆到AWS控制台，进入Athena服务，执行如下命令创建新表

CREATE EXTERNAL TABLE default.identity_center_users (
  Username string,
  UserId string
)
ROW FORMAT SERDE 'org.openx.data.jsonserde.JsonSerDe'
WITH SERDEPROPERTIES (
  'dots.in.keys' = 'false',
  'case.insensitive' = 'true',
  'mapping' = 'true'
)
LOCATION 's3://cur-lxy-us-east-1/user-list/'

由此在Athena的default数据库内，创建了一个名为identity_center_users的表。

创建表完毕后，查询测试：

SELECT * FROM "default"."identity_center_users" limit 10;

可看到返回结果正常。

#	username	userid
1	user01		843884c8-00c1-7046-e4e8-7ba5acc6cfc7
2	user02		04d86408-4001-7047-01d1-077885ab54fa

五、将CUR V2查询结果和用户映射表关联

以上两个步骤，分别有了Kiro用户使用Credit情况和Kiro用户名映射表两种类型的数据，由此可以拼接在一起。

1、查询所有用户的Credit使用情况（拼接用户名）

WITH user_plans AS (
    SELECT DISTINCT
        SPLIT_PART(line_item_resource_id, 'user/', 2) as user_id,
        CASE
            WHEN product_usagetype LIKE '%Pro+%' THEN 'Pro+'
            WHEN product_usagetype LIKE '%Power%' THEN 'Power'
            WHEN product_usagetype LIKE '%Pro%' THEN 'Pro'
        END as subscription_plan
    FROM "default"."cur-v2data"
    WHERE
        billing_period = '2026-01'
        AND line_item_product_code = 'Kiro'
        AND line_item_operation = 'monthly-subscription'
)
SELECT
    p.subscription_plan,
    CASE
        WHEN p.subscription_plan = 'Pro' THEN 1000
        WHEN p.subscription_plan = 'Pro+' THEN 2000
        WHEN p.subscription_plan = 'Power' THEN 10000
    END as plan_credits,
    u.username,
    SUM(c.line_item_usage_amount) as total_credits
FROM "default"."cur-v2data" c
JOIN user_plans p ON SPLIT_PART(c.line_item_resource_id, 'user/', 2) = p.user_id
LEFT JOIN "default"."identity_center_users" u ON p.user_id = u.userid
WHERE
    c.billing_period = '2026-01'
    AND c.line_item_product_code = 'Kiro'
    AND c.line_item_operation = 'Credits'
GROUP BY
    p.subscription_plan,
    u.username
ORDER BY
    CASE
        WHEN p.subscription_plan = 'Pro' THEN 1
        WHEN p.subscription_plan = 'Pro+' THEN 2
        WHEN p.subscription_plan = 'Power' THEN 3
    END,
    u.username

返回结果如下：

#	subscription_plan	plan_credits	username	total_credits
1	Pro	1000	user01	1.7210808227999999

2、查询超过Plan内Credit使用的用户（拼接用户名，如果没有则显示空）

WITH user_plans AS (
    SELECT DISTINCT
        SPLIT_PART(line_item_resource_id, 'user/', 2) as user_id,
        CASE
            WHEN product_usagetype LIKE '%Pro+%' THEN 'Pro+'
            WHEN product_usagetype LIKE '%Power%' THEN 'Power'
            WHEN product_usagetype LIKE '%Pro%' THEN 'Pro'
        END as subscription_plan
    FROM "default"."cur-v2data"
    WHERE
        billing_period = '2026-01'
        AND line_item_product_code = 'Kiro'
        AND line_item_operation = 'monthly-subscription'
),
user_usage AS (
    SELECT
        p.subscription_plan,
        CASE
            WHEN p.subscription_plan = 'Pro' THEN 1000
            WHEN p.subscription_plan = 'Pro+' THEN 2000
            WHEN p.subscription_plan = 'Power' THEN 10000
        END as plan_credits,
        p.user_id,
        SUM(c.line_item_usage_amount) as total_credits
    FROM "default"."cur-v2data" c
    JOIN user_plans p ON SPLIT_PART(c.line_item_resource_id, 'user/', 2) = p.user_id
    WHERE
        c.billing_period = '2026-01'
        AND c.line_item_product_code = 'Kiro'
        AND c.line_item_operation = 'Credits'
    GROUP BY
        p.subscription_plan,
        p.user_id
)
SELECT
    uu.subscription_plan,
    uu.plan_credits,
    u.username,
    uu.total_credits,
    uu.total_credits - uu.plan_credits as overage
FROM user_usage uu
LEFT JOIN "default"."identity_center_users" u ON uu.user_id = u.userid
WHERE uu.total_credits > uu.plan_credits
ORDER BY
    CASE
        WHEN uu.subscription_plan = 'Pro' THEN 1
        WHEN uu.subscription_plan = 'Pro+' THEN 2
        WHEN uu.subscription_plan = 'Power' THEN 3
    END,
    u.username

如果所有用户都没有超额使用，则查询结果为空。

六、参考文档

https://kiro.dev/docs/enterprise/monitor-and-track/dashboard/#dashboard-metrics

注：本文SQL由Kiro生成

-完-

最后修改于 2026-01-27