Anthropic SDK Token Usage Logger
Pass
Audited by Gen Agent Trust Hub on Mar 28, 2026
Risk Level: SAFE
Full Analysis
- [DATA_EXFILTRATION]: The skill transmits Anthropic API usage metrics, including token counts, latency, and calculated costs, to AWS CloudWatch and Datadog. These operations target well-known observability platforms and involve non-sensitive telemetry data.
- [COMMAND_EXECUTION]: Installation involves executing npx commands to download the skill from the author's repository (agentskillexchange/skills).
- [PROMPT_INJECTION]: The skill processes data from Anthropic API responses. Ingestion point: anthropic.messages.create output. Boundary markers: Not specified. Capability inventory: Network writes to AWS/Datadog and SNS notifications. Sanitization: Not specified. The risk is minimized as the skill targets structured usage metadata rather than free-form text.
Audit Metadata