analyze-costs
Pass
Audited by Gen Agent Trust Hub on Apr 1, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill utilizes structured Model Context Protocol (MCP) tools (
harness_get,harness_list,harness_execute) to fetch and interact with cloud cost data. - [SAFE]: Resource access is strictly limited to financial and cloud metadata resource types (e.g.,
cost_overview,cost_recommendation,cost_anomaly) consistent with the skill's stated purpose. - [SAFE]: The functionality is restricted to a specific environment (Harness MCP v2 server) as defined in the skill's metadata.
- [SAFE]: No hardcoded credentials, sensitive file access, or unauthorized network exfiltration patterns were observed.
- [SAFE]: The skill does not perform any remote code execution or package installation; it relies entirely on predefined tool calls provided by the agent's environment.
Audit Metadata