aws-cost-calculator-cli 1.0.3__tar.gz → 1.8.1__tar.gz

aws_cost_calculator_cli-1.8.1/CHANGES.md

@@ -0,0 +1,375 @@
+# Changelog
+
+## Version 1.8.0 (2025-11-07)
+
+### Added
+- **New `cc investigate` command** - Multi-stage cost investigation tool
+  - Combines cost analysis, resource inventory, and CloudTrail events
+  - Automatically finds SSO profiles for accounts
+  - Generates comprehensive markdown reports
+  - Supports `--no-cloudtrail` flag for faster execution
+  - Example: `cc investigate --profile myprofile --account 123456789012`
+
+### Changed
+- **Package cleanup** - Excluded backend, tests, and notebooks from PyPI distribution
+  - Backend code is only for Lambda deployment, not needed in CLI package
+  - Reduces package size and eliminates unnecessary files
+
+### Security
+- Verified no sensitive data (company names, account IDs) in PyPI package
+- All example data uses generic placeholders
+
+## Version 1.7.0 (2025-11-06)
+
+### Added
+- **`--resources` flag for `cc drill` command**: Resource-level cost analysis using CUR data
+  - Shows individual resource IDs (EC2 instances, RDS databases, S3 buckets, etc.)
+  - Queries AWS Cost and Usage Report via Athena
+  - Works with both local execution and Lambda backend
+  - Requires `--service` filter to be specified
+  - Example: `cc drill --service "EC2 - Other" --account 123456789012 --resources`
+  - Displays top 50 resources by cost with usage types and regions
+- **`cc setup-cur` command**: Configure CUR settings for resource-level queries
+  - Saves configuration to `~/.config/cost-calculator/cur_config.json`
+  - Prompts for: Athena database, table name, S3 output location
+  - Required before using `--resources` flag
+
+### Security
+- **CRITICAL**: Removed all hardcoded sensitive data from CUR queries
+  - No hardcoded account IDs
+  - No hardcoded database names
+  - No hardcoded S3 bucket paths
+  - All configuration must be provided by user via `cc setup-cur`
+
+### Technical Details
+- New module: `cost_calculator/cur.py` for local CUR queries
+- New module: `backend/algorithms/cur.py` for Lambda CUR queries
+- Updated `backend/handlers/drill.py` to support resource-level queries
+- Service-to-product-code mapping for flexible service name handling
+- Athena query execution with automatic polling and result parsing
+- CUR configuration loaded from `~/.config/cost-calculator/cur_config.json` or environment variables
+
+## Version 1.6.3 (2025-11-06)
+
+### Added
+- **`cc setup-api` command**: Automatically configure COST_API_SECRET
+  - Saves to `~/.zshrc` or `~/.bashrc` on Mac/Linux
+  - Saves to Windows user environment variables
+  - Prompts for secret with hidden input
+  - Updates existing configuration if already present
+
+## Version 1.6.2 (2025-11-06)
+
+### Fixed
+- Updated installation instructions to use PyPI instead of local editable install
+
+## Version 1.6.1 (2025-11-06)
+
+### Fixed
+- Sanitized all documentation to remove company-specific references
+- Removed real account IDs and credentials from examples
+- Updated all examples to use generic placeholder names
+
+## Version 1.6.0 (2025-11-06) - YANKED
+
+**Note:** This version was yanked due to unsanitized documentation containing company-specific information.
+
+### Added
+- **Flexible Authentication**: Multiple authentication methods for all commands
+  - `--sso <profile>`: SSO authentication with automatic login if session expired
+  - `--access-key-id`, `--secret-access-key`, `--session-token`: Static credentials
+  - Environment variable support: `AWS_PROFILE` for SSO, `AWS_ACCESS_KEY_ID` for static
+  - Auto SSO login: CLI automatically runs `aws sso login` if session is expired
+- Applied authentication flags to all commands: `calculate`, `trends`, `monthly`, `drill`
+
+### Changed
+- Enhanced profile loading to check `AWS_PROFILE` environment variable for SSO support
+- Updated README with comprehensive authentication documentation
+
+### Benefits
+- No manual SSO login required - CLI handles it automatically
+- Support for temporary credentials via CLI flags
+- Flexible authentication for different use cases (CI/CD, local dev, etc.)
+
+## Version 1.5.0 (2024-11-04)
+
+### Added
+- **Analyze Command**: Pandas-based aggregations and statistical analysis
+  - `cc analyze --type summary`: Aggregate costs across all weeks (sum, avg, std, min, max)
+  - `cc analyze --type volatility`: Identify high-volatility services and outliers
+  - `cc analyze --type trends`: Auto-detect increasing/decreasing cost trends
+  - `cc analyze --type search`: Filter services by pattern or minimum cost
+- **Profile Management**: CRUD operations for account profiles
+  - `cc profile list`: List all profiles
+  - `cc profile get --name <profile>`: Get profile details
+  - `cc profile create/update/delete`: Manage profiles in DynamoDB
+- **Lambda Enhancements**:
+  - New Analyze Lambda function with AWS Data Wrangler layer (pandas/numpy)
+  - New Profiles Lambda function for DynamoDB operations
+  - 2GB memory, 900s timeout for analyze function
+  - Time series analysis and statistical aggregations
+
+### Changed
+- Suppressed progress messages in JSON output mode for cleaner parsing
+- Updated API client to support new analyze and profiles endpoints
+
+### Fixed
+- JSON output now properly formatted without extra echo statements
+
+## Version 1.4.0 (2025-11-04)
+
+### Features Added
+
+#### Lambda API Backend Support
+- **Hybrid execution**: CLI now supports both local and Lambda API execution
+- Set `COST_API_URL` and `COST_API_SECRET` environment variables to use Lambda backend
+- Falls back to local execution if API not configured
+- **Identical results**: API and local execution produce exactly the same output
+- No changes to CLI interface - completely transparent to users
+
+#### Configuration
+- Environment variables:
+  - `COST_API_URL`: Base URL for Lambda API
+  - `COST_API_SECRET`: Shared secret for API authentication
+- Alternative: `~/.config/cost-calculator/api_config.json`
+
+#### Benefits
+- **Centralized logic**: Cost analysis runs on Lambda
+- **Multiple interfaces**: Same backend for CLI, web UI, Jupyter
+- **No code changes**: Existing CLI commands work identically
+- **Flexible deployment**: Choose local or API execution
+
+### Example Usage
+```bash
+# Use local execution (default)
+cc trends --profile myprofile
+
+# Use Lambda API
+export COST_API_URL="https://xxx.lambda-url.us-east-1.on.aws"
+export COST_API_SECRET="your-secret"
+cc trends --profile myprofile  # Same command, uses API
+```
+
+### Documentation
+- Added API client module
+- Added executor module for hybrid execution
+- Updated README with API configuration instructions
+
+---
+
+## Version 1.3.0 (2025-11-04)
+
+### Features Added
+
+#### Drill-Down Cost Analysis
+- **New command:** `cc drill --profile <name>`
+- Investigate cost changes at different levels of detail with automatic grouping
+- **Funnel approach**: Start broad, drill deeper based on findings
+- Supports three filter types:
+  - `--service`: Filter by service name (e.g., "EC2 - Other")
+  - `--account`: Filter by account ID
+  - `--usage-type`: Filter by usage type
+- Filters can be combined for deeper analysis
+
+#### Automatic Grouping Logic
+The command automatically shows the next level of detail:
+- **No filters** → Shows services (same as trends)
+- **Service only** → Shows accounts using that service
+- **Account only** → Shows services in that account
+- **Service + Account** → Shows usage types within that service/account
+- **All three filters** → Shows regions
+
+#### What It Shows
+- Week-over-week comparisons (default: 4 weeks)
+- Top 10 increases and decreases at the appropriate detail level
+- Total rows showing sum of top 10 changes
+- Percentage change and dollar amount
+- Filters out noise (>$10 and >5% change)
+- Most recent comparisons first
+
+#### Use Cases
+- **Investigate spikes**: See EC2 costs up? Drill to find which account
+- **Account analysis**: Which services are driving costs in a specific account?
+- **Service deep-dive**: Which usage types within EC2 are increasing?
+- **Root cause analysis**: Follow the funnel from service → account → usage type
+
+### Example Workflow
+```bash
+# 1. Start broad - see EC2 costs up $1000
+cc trends --profile myprofile
+
+# 2. Drill by service - see which accounts
+cc drill --profile myprofile --service "EC2 - Other"
+# Output: Account 123 is +$800
+
+# 3. Drill deeper - see usage types
+cc drill --profile myprofile --service "EC2 - Other" --account 123
+# Output: DataTransfer-Regional-Bytes is +$750
+```
+
+### Documentation
+- Updated README.md with drill-down examples and funnel approach
+- Added drill command to CLI help
+- Documented automatic grouping logic
+
+---
+
+## Version 1.2.0 (2025-11-04)
+
+### Features Added
+
+#### Month-over-Month Cost Analysis
+- **New command:** `cc monthly --profile <name>`
+- Analyzes month-over-month cost changes at service level
+- Compares consecutive calendar months (October vs September, etc.)
+- Generates comprehensive markdown report
+- Supports custom number of months: `--months 12`
+- JSON output option: `--json-output`
+- Custom output file: `--output monthly_trends.md`
+
+#### What It Shows
+- Service-level aggregation (total cost per service)
+- Top 10 increases and decreases per month comparison
+- Total rows showing sum of top 10 changes
+- Percentage change and dollar amount for each service
+- Filters out noise (>$50 and >5% change)
+- Most recent comparisons first (reverse chronological)
+
+#### Use Cases
+- Monthly cost review and budget tracking
+- Identifying large month-over-month changes
+- Understanding seasonal cost patterns
+- Tracking major service additions or removals
+- Long-term cost trend analysis
+
+### Example Output
+```
+October 2025 → November 2025
+  Increases: 1, Decreases: 10
+  Top: Savings Plans for AWS Compute usage (+$231,161.46)
+```
+
+### Documentation
+- Updated README.md with monthly command examples
+- Added monthly report documentation
+
+---
+
+## Version 1.1.0 (2025-11-04)
+
+### Features Added
+
+#### Cost Trends Analysis with Dual Methodology
+- **New command:** `cc trends --profile <name>`
+- Analyzes cost changes with **two comparison methods**:
+  1. **Week-over-Week (WoW)**: Compares consecutive weeks - catches immediate spikes
+  2. **Trailing 30-Day (T-30)**: Compares to 4 weeks ago - shows sustained trends
+- **Service-level aggregation**: Shows total cost per service (not usage type detail)
+- Generates comprehensive markdown report with both methodologies
+- Supports custom number of weeks: `--weeks 18`
+- JSON output option: `--json-output`
+- Custom output file: `--output weekly_trends.md`
+
+#### What It Shows
+- **Week-over-Week section**: Consecutive week comparisons (Monday to Sunday)
+- **T-30 section**: 30-day baseline comparisons
+- Top 10 cost increases and decreases for each comparison
+- **Total rows**: Sum of top 10 changes for quick assessment
+- Percentage change and dollar amount for each service
+- Automatically filters out noise (>$10 and >5% change)
+- Most recent comparisons first (reverse chronological)
+
+#### Why Two Methodologies?
+- **WoW**: Catches immediate changes, spikes, and weekly volatility
+- **T-30**: Filters out noise, reveals sustained trends and real cost changes
+- Together they provide both **immediate alerts** and **trend analysis**
+
+#### Use Cases
+- Weekly cost monitoring and anomaly detection
+- Identifying sudden cost spikes (WoW) vs sustained increases (T-30)
+- Understanding which services are trending up or down
+- Tracking cost optimization efforts over time
+- Distinguishing temporary spikes from permanent cost changes
+
+### Example Output
+```
+WEEK-OVER-WEEK:
+  Week of Oct 19 → Week of Oct 26
+    Increases: 4, Decreases: 10
+    Top: EC2 - Other (+$949.12)
+
+TRAILING 30-DAY (T-30):
+  Week of Oct 26 vs Week of Sep 28
+    Increases: 10, Decreases: 10
+    Top: EC2 - Other (+$886.39)
+```
+
+### Documentation
+- Updated README.md with dual methodology explanation
+- Added trends report format documentation
+- Updated CLI help text
+
+---
+
+## Version 1.0.0 (2025-11-04)
+
+### Features Added
+
+#### 1. Static AWS Credentials Support
+- **New command:** `cc configure --profile <name>`
+- Allows using static AWS credentials instead of SSO
+- Supports temporary credentials with session tokens
+- Credentials stored securely in `~/.config/cost-calculator/credentials.json` (600 permissions)
+
+#### 2. Dual Authentication Methods
+- **SSO Method:** For interactive use (existing)
+  ```bash
+  aws sso login --profile my_sso_profile
+  cc calculate --profile myprofile
+  ```
+
+- **Static Credentials Method:** For automation/CI (new)
+  ```bash
+  cc configure --profile myprofile
+  # Enter: Access Key ID, Secret Access Key, Session Token (optional)
+  cc calculate --profile myprofile
+  ```
+
+#### 3. Enhanced Help Documentation
+- Updated `cc --help` with authentication method examples
+- Added comprehensive guides:
+  - `COST_CALCULATION_METHODOLOGY.md` - Formula explanation
+  - `WEEKLY_COST_HISTORY.md` - Historical cost trends
+  - `PYPI_UPLOAD_GUIDE.md` - Publishing instructions
+
+#### 4. PyPI Packaging
+- Prepared for PyPI distribution as `aws-cost-calculator-cli`
+- Added proper metadata, classifiers, and dependencies
+- Created upload script: `./upload_to_pypi.sh`
+- Added LICENSE (MIT), MANIFEST.in, .gitignore
+
+### Security Enhancements
+- Credentials file permissions set to 600 (owner read/write only)
+- Sensitive files excluded from git (.gitignore)
+- No credentials leaked in error messages or logs
+
+### Documentation
+- **README.md:** Installation and usage guide
+- **COST_CALCULATION_METHODOLOGY.md:** Detailed formula explanation
+- **WEEKLY_COST_HISTORY.md:** 18 weeks of historical data
+- **PYPI_UPLOAD_GUIDE.md:** Publishing instructions
+
+### Commands Available
+1. `cc init` - Initialize profile with account list
+2. `cc configure` - Configure AWS credentials (NEW)
+3. `cc calculate` - Calculate costs
+4. `cc list-profiles` - List configured profiles
+
+### Configuration Files
+- `~/.config/cost-calculator/profiles.json` - Profile configurations
+- `~/.config/cost-calculator/credentials.json` - AWS credentials (NEW)
+
+### Next Steps
+1. Test with temporary AWS credentials
+2. Upload to PyPI
+3. Install from PyPI: `pip install aws-cost-calculator-cli`

aws_cost_calculator_cli-1.8.1/MANIFEST.in

@@ -0,0 +1,3 @@
+include README.md
+include LICENSE
+include CHANGES.md