creating-surveys
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [SAFE]: The skill implements a secure-by-default workflow. It instructs the agent to create surveys as drafts and perform a comprehensive review with the user (using
posthog:survey-get) before proceeding to the launch step (posthog:survey-launch). - [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided text to populate survey configurations. It mitigates the risk of malicious content by requiring the agent to present the final questions, audience targeting, and delivery methods to the user for validation (boundary marker/human review). The agent's capabilities are scoped to specific survey management tools.
- [DATA_EXFILTRATION]: The instructions include a specific safeguard regarding data privacy, informing the agent that survey content is public. It explicitly directs the agent to avoid copying private customer information into these fields, which prevents accidental data exposure through the public survey platform.
Audit Metadata