cursor-cost-optimization
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests usage data from the
cursor-usageMCP server, creating a potential surface for indirect instructions hidden in usage metadata such as user names or billing group identifiers. However, the skill's toolset is restricted to cost-management functions and lacks high-privilege capabilities such as arbitrary shell execution or network exfiltration, making the risk negligible in this context. - [EXTERNAL_DOWNLOADS]: The documentation includes a reference to an external GitHub repository (
github.com/ofershap/cursor-usage-tracker) for extended monitoring and anomaly detection. This repository is owned by the skill author and is presented as a manual resource for the user to explore, with no evidence of automated or malicious execution patterns.
Audit Metadata