shield_lock
金大哥
← 返回 Skill 列表
extension
分类: 开发与工程API Key 暂未确认

aws-billing-report

生成专业的AWS账单Excel报表，包含与AWS账单页面完全匹配的详细月度成本细分，包括数据传输和所有使用类型。当用户请求AWS账单报告、成本导出、月度成本报告或提及“aws billing”、“aws费用”、“billing report”、“kolya billing”、“生成AWS账单报表”、“导出AWS费用”等关键词时使用。使用成本和使用情况报告（CUR）以获取最详细的账单数据。

person作者: jakexiaohubgithub
open_in_new仓库download下载资源包

AWS Billing Report

Overview

Generate professionally formatted Excel reports with detailed AWS billing data that matches the AWS Bills page exactly:

  • Sheet1: Service-level summary with monthly costs
  • Sheet2: Detailed drill-down with three-level hierarchy (Service → Region → Usage Type)
  • Complete breakdown including Data Transfer, EC2 instances, and all usage types
  • Year-over-year comparison with growth percentages
  • Microsoft 365 professional styling (Aptos font, light gray headers)

Data Source: Cost and Usage Report (CUR) - provides the most detailed billing information available from AWS.

Platform Detection

IMPORTANT: Before starting, check if you are running inside the OpenOps platform by looking for these signals in your system prompt:

  • "You are OpenOps AI" or "CRITICAL ENVIRONMENT RULES"
  • "Available Cloud Accounts" section with account IDs and regions
  • A WORKSPACE path is specified

If running inside OpenOps → Auto mode

When inside OpenOps, most parameters are already available:

  • Auth: Use default credential chain — credentials are injected via environment variables. Do NOT use --profile.
  • Accounts: Extract from the "Available Cloud Accounts" section in the system prompt.
  • Output path: Use the WORKSPACE path from system prompt: <WORKSPACE>/<YYYY-MM-DD>/aws_billing_<account>_<year>.xlsx. Create the date directory first with mkdir -p.
  • Script path: Skills are symlinked at .claude/skills/aws-billing-report/scripts/ relative to the working directory.

Ask the user for the following before proceeding:

  1. Which account to generate the report for (if multiple accounts available)
  2. Which year (default: current year)
  3. Which month(s) — present options: specific month (e.g. "3" for March), a range (e.g. "1-3" for Jan–Mar), or "all" for the full year

CRITICAL: You MUST wait for the user's response before proceeding. Do NOT assume defaults for month selection. This is a mandatory interaction point.

Then proceed to Step 2.

If running standalone → Interactive mode

Ask the user for the following information if not provided:

  1. AWS Account ID (required): 12-digit AWS account ID
  2. AWS Profile (optional): --profile <name>, or uses default credentials
  3. Year for the report (default: current year)
  4. Month(s) — specific month, range, or "all"
  5. Output path for the Excel report

CRITICAL: You MUST wait for the user's response before proceeding to Step 2. Do NOT run fetch and report generation in the same turn as the questions.

Then proceed to Step 2.

Workflow

Step 1: Resolve Script Path

Resolve SCRIPT_PATH: The scripts live in the scripts/ directory where this SKILL.md is located:

  • OpenOps mode: SCRIPT_PATH=".claude/skills/aws-billing-report/scripts"
  • Standalone mode: Use the absolute path, e.g. SCRIPT_PATH="$HOME/.agents/skills/aws-billing-report/scripts" or wherever installed.

Step 2: Check and Setup CUR

Check if CUR is already configured:

python3 <SCRIPT_PATH>/setup_cur.py --check-only [--profile <PROFILE>]

If CUR is configured: → Proceed to Step 3.

If CUR is NOT configured:

python3 <SCRIPT_PATH>/setup_cur.py [--profile <PROFILE>] [--account-id <ACCOUNT_ID>]

Setup creates: S3 bucket aws-cur-{account_id}, CUR report definition (Parquet, daily).

Important: ⏰ First CUR report takes 24 hours to generate. Inform the user and stop if this is the first setup.

Step 3: Fetch Billing Data

python3 <SCRIPT_PATH>/fetch_aws_billing.py <ACCOUNT_ID> <YEAR> --months <MONTHS> [--profile <PROFILE>] > billing_<YEAR>.json 2>/dev/null

IMPORTANT: The script outputs JSON to stdout and progress logs to stderr. Always redirect stdout to file and discard or separate stderr. Never use 2>&1 — it mixes logs into the JSON and corrupts the output.

--months options: single (3), range (1-6), comma-separated (1,3,6), or all. Default: current month.

For multi-year reports, fetch each year separately.

Step 4: Generate Excel Report

python3 <SCRIPT_PATH>/generate_excel_report.py <OUTPUT_PATH> billing_<YEAR1>.json [billing_<YEAR2>.json ...]

The generated report contains:

Sheet1 - Service Summary:

  • Service-level costs for each month
  • Clean service names matching AWS Bills page
  • Total row with auto-calculated formulas

Sheet2 - Usage Details:

  • Three-level drill-down: Service → Region → Usage Type
  • Description, Quantity with unit, Rate with unit, Cost

Step 5: Deliver Report and Cleanup

  1. Verify the file was created successfully
  2. Display a summary (total cost, top services, service count)
  3. Clean up temporary JSON files: rm -f billing_*.json

Prerequisites

Automatically handled during execution:

  1. Python Packages: boto3, pyarrow, pandas, openpyxl — install if missing: 
    pip install boto3 pyarrow pandas openpyxl
  2. AWS Permissions: S3 (CreateBucket, PutBucketPolicy, ListBucket, GetObject), CUR (PutReportDefinition, DescribeReportDefinitions)

Error Handling

  • Missing credentials: Check profile, suggest aws sso login
  • CUR not configured: Auto-run setup, inform about 24-hour wait
  • Missing packages: Display pip install command
  • No CUR data: Check if 24 hours passed since setup