configuring-experiment-rollout
Pass
Audited by Gen Agent Trust Hub on Apr 30, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is purely instructional and defines how an agent should handle user requests for experiment configuration. It uses a structured approach to resolve ambiguity by asking the user for clarification before proceeding.
- [EXTERNAL_DOWNLOADS]: The skill includes links to official PostHog documentation for further reading. These links target the vendor's own domain and do not involve executable code downloads or untrusted sources.
- [SAFE]: No evidence of malicious patterns such as prompt injection, hardcoded credentials, obfuscation, or unauthorized data access was found. The instructions prioritize data integrity and user experience by requiring warnings for non-standard configurations.
Audit Metadata