- Waste identification: Find unused columns, never-queried datasets, and high-volume low-value data
- Query analysis: Analyze query patterns to identify optimization opportunities
- Dashboard deployment: Deploy cost control dashboards with ingest trends, burn rate, and projections
- Monitor creation: Set up monitors for contract limits, spike detection, and reduction glidepaths
- Analysis workflow: Systematic approach to prioritize and analyze datasets by impact
Prerequisites
- Create an Axiom account.
- Create a dataset in Axiom where you send your data.
- Create an API token in Axiom with permissions to ingest data to the dataset you have created.
Install control costs skill
Choose one of the following ways to install the control costs skill:- Quick install
- Amp
- Claude
- Claude Code
- Cursor
- Other agents
Install all Axiom skills at once for Claude Code, Cursor, and other Claude-compatible agents:
Configure Axiom credentials
All Axiom Skills share the same credential configuration. Create a configuration file at~/.axiom.toml:
~/.axiom.toml
Replace
API_TOKEN with the Axiom API token you have generated. For added security, store the API token in an environment variable.Replace ORGANIZATION_ID with your organization ID. For more information, see Determine organization ID.For token creation and scoping guidance, see Token hygiene for AI agents.Use control costs skill
The control costs skill activates automatically when you ask your AI agent to:- Reduce Axiom costs or find waste
- Set up cost control dashboards and monitors
- Analyze query coverage and unused data
- Track ingest spend and burn rate
- “Help reduce Axiom costs”
- “Find unused columns in the datasets”
- “Set up cost control monitoring for the organization”
- “Which datasets are being paid for but never queried?”
Cost control workflow
The control costs skill follows a phased approach:| Phase | Description |
|---|---|
| Discovery | Capture baseline ingest statistics and build analysis queue |
| Dashboard | Deploy cost control dashboard with trends and projections |
| Monitors | Create monitors for contract limits and spike detection |
| Optimization | Analyze datasets to find waste and recommend actions |
| Glidepath | Track reduction progress with weekly threshold updates |