For the complete documentation index, see llms.txt. This page is also available as Markdown.

Cost Management

Oscar provides cost tracking and analysis for your cloud infrastructure. Configure cost data sources, monitor warehouse status, and get AI-powered spending insights.

Setup Cost Tracking

Configure cost data sources with /cost setup:

/cost setup

Supports:

  • AWS - Cost and Usage Reports (CUR)

  • Azure - Cost Management exports

The setup wizard guides you through selecting your cloud provider, configuring export locations, setting up access credentials, and validating the configuration.

AWS setup:

AWS cost setup

Azure setup:

Azure cost setup

Check Warehouse Status

View your cost data warehouse status with /cost status:

Cost warehouse status

Shows:

  • Configured cost sources

  • Last sync time

  • Data freshness

  • Available date ranges

  • Any sync errors or warnings

Get Cost Summary

Analyze spending patterns with /cost summary:

Oscar analyzes your cost data and provides month-over-month spending trends, top cost drivers by service, unusual spending patterns, and cost optimization opportunities.

Cost breakdown by service:

See your spending organized by cloud service with month-over-month comparisons and percentage changes.

Cost summary table

Flagged findings:

Oscar identifies unusual spending patterns, cost spikes, and potential issues that need attention.

Cost flagged findings

Next steps:

Get actionable recommendations for cost optimization and investigation priorities.

Cost next steps

The summary is AI-generated based on your actual cost data, providing context-aware insights specific to your infrastructure.

Cost Data Warehouse

Oscar uses DuckDB to create a local cost data warehouse. Cost data is:

  • Synced from your cloud provider's cost exports

  • Stored locally in ~/.config/oscar/cost/

  • Queried for analysis and reporting

  • Never sent to external services (runs entirely locally)

Notes

  • Cost data requires initial setup via /cost setup

  • First sync may take several minutes depending on data volume

  • Data is refreshed automatically when you run /cost status or /cost summary

  • Historical data availability depends on your cloud provider's export retention

  • All cost analysis runs locally using DuckDB

Last updated

Was this helpful?