exploring-llm-costs
Pass
Audited by Gen Agent Trust Hub on Apr 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill utilizes official PostHog tools for data querying and visualization, such as
posthog:execute-sql,posthog:insight-create, andposthog:alert-create, which are appropriate for its stated purpose. - [SAFE]: All external links and documentation references point to legitimate vendor-controlled domains, specifically
posthog.comandapp.posthog.com, posing no risk of unauthorized external communication. - [SAFE]: The provided SQL (HogQL) templates include explicit type casting for event properties (e.g.,
toFloat,toInt), which follows best practices for data handling and provides a basic layer of sanitization for numeric values. - [SAFE]: No malicious patterns such as prompt injection, hardcoded credentials, obfuscation, or unauthorized command execution were detected across any of the skill's files.
Audit Metadata