creating-ai-subscription
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill uses user-supplied free-text prompts to drive an AI-based report generation pipeline. \n- Ingestion points: The
promptargument in theposthog:subscriptions-createtool. \n- Boundary markers: The instructions do not specify any delimiters to separate the user prompt from the system instructions. \n- Capability inventory: The user input is used by the backend to generate and execute HogQL queries and synthesize the final report. \n- Sanitization: No specific validation or sanitization protocols for the user-supplied prompt are mentioned in the skill instructions.
Audit Metadata