configuring-experiment-analytics
Pass
Audited by Gen Agent Trust Hub on Nov 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements standard administrative tasks for PostHog experiment configuration. It includes clear instructions on using platform-provided tools for data retrieval and updates.
- [INDIRECT_PROMPT_INJECTION]: The skill processes project-specific data (event names) provided by users or the project schema. It mitigates potential injection risks by mandating a schema discovery step via
read-data-schemaand requiring user confirmation before incorporating these names into experiment configurations, ensuring that only valid and expected event names are used.
Audit Metadata